PRE2019 3 Group7

From Control Systems Technology Group

Jump to: navigation, search

Contents

Group Members

Name Student ID
Daan Schalk 0962457
Job Willems 1003011
Jasper Dellaert 1252454
Sanne van Wijk 1018078
Wietske Blijjenberg 1025111

Introduction

Problem statement

When looking at Stratumseind these days, clubs and bars that were once flourishing look empty and silent. For a manager of such a bar, it may be difficult to decide what to improve to regain the glory of those golden days. And, maybe more importantly, how to make profit again. Humans are complex, so their happiness and willingness to spend depends on different factors combined in a specific way to find a balance between keeping the customers happy and getting enough money. Because of all these different factors, experimenting with them in real life is impossible work. On top of that, change in revenue might not be seen in a week, or even multiple weeks. To be sure about the effect of a factor, a longer observation period is needed. However, if you need to test a lot of different factors in a lot of different combinations, this can take ages. Which a struggling bar owner will not have.

That is where a simulation of the behaviours of the people in a bar or club might help. The simulation will make it easy to experiment with different factors, and observe the effect of those factors on the behaviour of the people in the bar. It also adds the possibility to run multiple simulations in sequence. This way, certainty can be reached about the influence of a certain factor on the revenue, and managers of clubs and bars can reach decisions about what to improve to optimize their revenue. The average club derives around two-thirds of its revenue from the sale of beverages, and it is therefore a vital source of profits [1]. Therefore the simulation will be focusing on how to maximize the selling of beverages.

Objectives

  • Main goal: assist the owner of a bar or dance hall with ideas on how to improve customer revenue.
  • Construct a multi-agent simulation with which we can find what variables with what capacities are needed to maximize generated income.
  • Analyse the reliability of such a simulation by discussing the observations of the client and the results of previous research done in this field (if any apply).
  • Analyse the possibilities and shortcomings of such a simulation.

State-of-the-art

AI and simulations

Artificial Intelligence (AI) is used more and more in (environmental) modelling. It can mimic human perception, learning and reasoning, which can be used to solve complex problems. There are a lot of difference AI techniques: case-based reasoning, rule-based systems, artificial neural networks, genetic algorithms, cellular automata [2], fuzzy models, multi-agent systems, swarm intelligence, reinforcement learning, hybrid systems, and, arguably, Bayesian networks and data mining [3]. Not all of those techniques seem fit for modelling a simulation of a bar environment: for case-based reasoning, similar past cases are needed which are used to produce a solution to the current problem [4] [5]. However if there are no similar past cases, like in this case, CBR cannot draw inferences about the problem. On top of that, CBR is a black-box approach and offers little insight into the system and processes involved [3]. Since process explanation would be quite useful in this simulation, other black-box approaches like Artificial Neural networks [6] [7] [8] or genetic algorithms [9], a search technique mimicing natural selection [10] are not ideal either.

Reinforcement learning can be discarded as well. It learns through interaction between a learning agent and the environment, using trial-and-error to achieve a goal[11]. It is particularly useful in robotics and game playing [12], where it creates new behaviour rather than modelling existing behaviour. However, in this project modelling existing bahaviour is desired, since it is the goal to find out how people behave in certain settings. On top of that, RL alone is often not enough to solve other problems. RL in combination with other AI techniques is useful, but it is difficult to formulate a policy that works successful in real life [13].

A rule-based system could be used, which solves problems by rules derived from expert knowledge [14]. The rules consist of an if-part and a then-part: if (certain condition) then (do this action). These rules are fed to an inference engine, which has a working memory of information about the problem, a pattern matcher and a rule applier. The pattern matcher consults the memory to decide which rules are relevant, after which the rule applier chooses what rule to apply [3]. The then-part of the rule often creates new information, which is added to the working memory. This cycle is repeated until no more relevant rules are found [15]. This is easy to understand, implement, and maintain [3]. However, an RBS requires comprehensive knowledge to implement. Also, a RBS does not seem fit for an environment with multiple people, unless every person is its own rule-based system.

Something more fit would be a multi-agent system, which consists of a network of agents which interact to achieve goals [16]. Each agent is a software component which contains code and data [17]. Each agent on its own is incapable of solving the problem assigned to the MAS, but all agents together have the potential to solve it [18]. The agents can communicate using a high-level Agent Communication Language (ACL) through which they share information, request services and negotiate with each other [17]. In the context of using a simulation to optimize the revenue of a pub, the patrons of the pub would be the agents. A MAS is great to model complex systems with multiple interactions among dynamic and autonomou entities [3].

However, the effectiveness of a MAS depends largely on the agent organisation. When using peer-to-peer infrastructure, network maintenance can be problematic [3].

With a crowd in mind, a swarm intelligence maybe seems like a good approach. This form of agent-based modelling is inspired by colonies of social animals such as bees and ants [19] or schools of fish. This is interesting because simple individual agents can exhibit higher intelligence as a swarm. Local interactions can let global patterns emerge, without centralised control or a global model [3]. Algorithms for SI techniques are versatile yet easy to implement. The group of agents has a self-organisation which allows adaptation to changes in the environment, which makes a SI system robust against failures and perturbations [20]. This gives them the ability to solve dynamic problems as well [3].

A hybrid system could also be considered. In a hybrid system, two or more AI techniques are combined to overcome weaknesses presented by both techniques when used on their own. There are three main types of hybrid systems [21] :

  • Sequential: the first technique passes its output to the second to generate the output.
  • Auxiliary: the first technique obtains some information from the second to generate the output
  • Embedded: the two techniques are contained within one another.

The most common hybrid is a neuro-fuzzy system, which combines an ANN and a fuzzy system. A fuzzy system uses fuzzy sets to deal with imprecise and incomplete data. Fuzzy set membership differs from normal set membership in that it takes a value between 0 and 1, instead of being true or false. This enables fuzzy models to describe vague statements as in natural language [22]. Exact input values are transformed into fuzzy memberships through a process called fuzzification [23]. The model is then built on prior rules combined with fuzzified data by the fuzzy inference machine. The fuzzy output is then transformed to a crisp number, which is called defuzzification [24]. It has a great ability to handle vague or imprecise information[3], however it is mostly more difficult to understand and apply than other AI techniques. Good membership functions are hard to determine. Also, fuzzy systems have no learning capability or memory [25]. Pairing it with an ANN creates a fast, efficient and easily designed, implemented and understood type of AI[26].

In this situation, a multi-agent based system paired with a rule-based system could be nice, where each agent operates according to a certain set of rules which determine its actions. The most important thing needed to realize this, is comprehensive knowledge of the situation and the patrons, which should be achieved using elaborate research.

Optimizing the revenue of a bar

There has not been a lot of research into how to optimize the revenue of a bar, it is mostly done on experience. There has however been one previous study using an agent-based approach to optimize bar arrangements [27]. It considers the balance between income and customer-satisfaction, reasoning that while more visitors is great for the revenue of a bar, more visitors also make it more crowded which is unpleasant for the visitors. Thus it can be difficult to decide a maximum number of people allowed in the venue. The authors propose a model that simulates a crowd to find the optimal arrangement, which is defined as the placement of the bars, for a concert-like venue. The model used is agent-based. The agents are individual humans with a beer-need, and one of two different modes: stage-mode and bar-mode. When they are in stage-mode, the point of interest is the stage. In bar-mode, the point of interest is the bar. The crowd-dynamics emerges from the agent’s individual behaviour, reminding one of a swarm intelligence [19]. Although this model was effective, which confirms that an agent-based model would be a valid choice, it is mentioned that it is far from perfect for explaining real-life events. Especially the movement of the agents was not quite right. This seems like an area to improve in this project.

Approach

1.Determine important variables for simulation

1.1 Determine what is important from the perspective of a pub owner
1.2 Determine what is previously described as important in literature

2. Research

2.1 Determine correlations between the determined variables and alcohol consumption trough literature study.
2.2 Formalyze hypothesis etc.

3. Determine unknowns in correlations, research those correlations.

3.1 Set up a research plan.
3.2 Execute the research plan.
3.3 Analyze the results.

4. Set up a simulation using the found correlations.

4.1 Create a minimum viable product: a bar setting with AI using that bar.
4.2 Start implementing each of the correlations found in 2 and 3.

5 Analyze the simulation

5.1 Change variables, optimize the simulation.
5.2 Refer back to research and hypothesis, is our simulation realistic and how does it comply with our hypothesis?
5.3 If neccecary review steps 2, 3 and 4.

6 Finalize wiki and conclusions

Users

The main stakeholder for this simulation are bar/club owners. They will be able to specify what their establishment currently looks like, and what they are willing to adjust in their establishment. The simulation will then be able to find out how to maximize profit in the specified establishment. Since the simulation takes into account which factors a bar/club owner has influence on, it can be a great tool for giving meaningful suggestions to generate more profit. It can also be used by a bar/club owner to see if an idea he has to generate more profit has potential or not, since all configurations of factors can be tested.

Since it focuses mainly on how to maximize the sale of alcohol, it could also be useful in other contexts where selling alcohol is the main way of making profit, as long as the setting itself does not differ too much from the bar setting specified in the simulation. After all, differing from the defined setting could possibly introduce factors which change the result of the simulation. It could be decided per case if the simulation would be useful, or if too many unknowns are introduced in a certain situation. The simulation could even be useful in settings like student parties, where people want to get others as drunk as possible. Question is if it is ethical to use such a tool to get people drunk.

In certain cases, it could even be used to predict the amount of alcohol that will be sold on a certain day. This could be useful when deciding how much alcohol to order. Of course experienced bar/club owners will already have a feeling for how much alcohol to buy, but especially in the case of student parties, it is not always known how much should be bought.

A secondary user for which this simulator might prove useful, are the people going to the bars and clubs. Assuming that customers spend more money when they enjoy themselves and leave when they are unhappy, using the results of the simulator to improve bars would mean that bars become more enjoyable. This is beneficial for the customers: they will have a better time.

Requirements & constraints

It is important that the resulting simulation is actually useful for the users. Therefore a list of requirements based on the user specification and input of the user is specified as follows:

Requirements

  • The simulation must be able to house multiple agents (> 5)
  • The simulation must accurately simulate a bar environment
  • The simulation must accurately simulate the influence of environmental factors on the agents (reflects the real world)
  • The simulation must accurately simulate the influence of the agents on each other (reflects the real world)
  • The simulation must accurately simulate the actions of patrons in a bar environment (reflects the real world)
  • The output of the simulation must show how to optimize the revenue in the simulated bar (it must be possible to make conclusions about what factors to change)
  • The output of the simulation must consider factors that a pub owner can influence

Preferences

  • The simulation could have a visualization of the bar environment and the agents

Constraints

Constraints are limits that are set for the simulation. Constraints are needed in order to create a working simulation. Without these constraint a simulation would prove too complex to make, therefore these constraints have been made.

Bar

  • Bar cannot be remodeled or renovated
  • There is only one type of alcohol/drink
  • Other bars/competitors are not taken into account
  • Given one night with certain amount of people inside
  • Bar will not sell food
  • No animals are allowed
  • Drinks will not expire
  • The bar only has one room in which people will dance and drink

People

  • People will not start fights
  • People will not complain
  • People will not break stuff
  • People will always pay and not steal
  • People will go when they are unhappy or when they have no more energy
  • People do not go to bathroom

Research

Input stakeholders

Of course it is important to know what the stakeholders want and what is and is not possible in a bar setting. Therefore some people knowledgeable of the inner workings of a bar were asked a few questions about the factors that could have impact on the revenue of a bar. Most of them were certain of one thing: drink discounts, special actions (for different nights) and promotions increase profit. Others also named theme evenings and hosting parties as a good way to make profit. It was also mentioned that giving customers the possibility to pay by card increased profits. This sounds logical: previously, people would take a certain amount of cash with them, and if they spend it all, they were done. With the introduction of the possibility to pay by card, the amount of money people can spend at a bar gets more limitless: it is, in theory, only limited by the amount of money on your bank account.

When asked what people could change and could not change, a lot of possibilities and restrictions were named. Most cafes can change or have previously changed the music volume, music genre, drink specials, personnel (for example hire people with charisma), lighting, painting, seating and tables, although people mentioned that a lot of interior design is impossible due to fire safety issues. People did not mind changing these things. The size of the café could never be changed, as well as the location of the restrooms, the opening hours, the admission policy for people younger than 18, the maximum sound volume, the smoking policy, the location of the bar, beer prices and lighting setup. In some bars, for example a karaoke bar, different music was not really a possibility.

An interesting thing mentioned that people did not want to change, was the atmosphere. Some pubs were not interested in making profit if it would mean their pub would not be 'their' pub anymore.

Having an indication of what pub owners would be willing to change in their establishment, the next step is checking what influence those possible changes could have on the patrons.

Bar-related factors

Music

Music can have quite an influence on how people behave in social settings. Most people like background music: studies have found that people spend less time drinking in bars that don’t play music [28]. But there are many different genres of music, and many different ways to play music. The two most important aspects to music are volume and genre.

Volume

Of course bar owners can not turn the volume up past the by law specified maximum volume, but they can still choose between playing music softly, as background music, or playing music louder. It appears to be important for alcohol sales to choose the correct volume: according to a previous study, loud music makes people drink 31% more [29] [30] [31]. The reason behind this is could be that when people can not communicate due to the noise in the bar, they start focussing on drinking [32]. Another explanation is that high sound levels may cause higher arousal, which leads the subjects to drink faster and order more drinks [29] [33]. This suggests that bar owners should not be afraid to turn up the music.

Genre

When not running a karaoke bar, pub owners have quite a say in what genre music should play. They can choose a certain genre, mix all genres, just play hit songs, take requests, not take requests, you name it. Genres can have quite different tempos [34], and there can even be a lot of difference in tempos between music from the same genre. It is important to consider this: a lot of studies have found all kinds of effects of the tempo of music on the behaviour of people. One study links faster music to faster drinking [35] [36]. However, a different study links slower music to more sales. Customers stayed about the same amount of time but spent more during slow music than during normal or fast music. [37]. An explanation for this difference is that people prefer to listen to music that moderates their state of arousal [38]. During a relaxing activity like an afternoon beer, people prefer slow low-arousal music, while during a night out people prefer music that further heightens their state of arousal. For a pub this means that the tempo of the music should correspond to the state of arousal of the patrons: if they are more aroused, they need faster music, if they are less aroused, they need slower music.

On top of the tempo of the genre, studies found that the specific genre of music matters. Jacob (2006) found that when playing drinking songs in a bar, the duration of stay and spending both are increased [39]. Further research supported this and claimed that customers exposed to textual references to alcohol spent significantly more on alcoholic drinks than those who were not [40].

It was also found that in social context, people modify their bodily behaviour according to the dynamic level of the bass drum. More specifically, they move more actively and display a higher degree of tempo entrainment as the sound pressure level of the bass drum increases [41]. This could be interesting if a correlation between dancing and alcohol could be found.

Drink specials

The alcohol prices are important for choosing a bar, but not as much as other factors. People like everyday low alcohol prices but are prepared to pay a bit more, as long as it is not too much [42]. Low alcohol prices do have a significant effect on alcohol intake however: another study found that a happy hour with price reduction increases alcohol consumption. When the purchase price was reduced by half, casual and heavy drinkers increased their consumption eight and nine times respecively [43].

Women are more likely to take advantage of drink specials, whereas men reported greater alcohol expenditures, consumed more drinks, and drank for longer periods of time. Participants in bar-sponsored drink specials drank more. [44]

Lighting

A lot of studies have been done on the impact of lighting on people. Lighting can influence how people perceive an environment, which in turn impacts their actions and behaviour. The common visiting hours of pubs and cafes are in the evening, therefore a pub relies almost entirely on artificial lighting to create the correct ambience. This gives the pub owners a lot of power: they can adjust the lighting almost entirely according to their own wishes, as long as the lighting setup does not have to change. Studies about lighting often consider two factors: brightness and colour.

Colour

Red is perceived as negative and tense as well as physically arousing. Blue has been identified as calm, cool, and positive. More positive retail outcomes occurred in a blue rather than a red environment [45]. In general, cool colours were found to be less arousing than warm colours. Pleasure and dominance were not affected by the warmness of the colour [46]

People felt more pleasure when colour and scent were combined based on arousal congruence [46]. This suggests that when designing a cafe, picking values for different factors with congruent arousal improves happiness of the patrons.

This study found that under white and green lighting a space is perceived as more useful, spacious, clear, and luminous than under red lighting. Green and white lighting were perceived equally comfortable in an interior space. Chromatic coloured lighting was perceived to be more aesthetic than white lighting.

Brightness

Some people may recognize the feeling of still feeling drowsy in the morning until you open up the curtains and the bright rays of the sun blaze you awake. This is not just in your mind: some studies suggest that brightness increases cognitive alertness and enhances self-control, whereas darkness can cause cognitive dullness and reduces self-control [47] [48] [49] [50] [51]. Darkness can also increase cognitive alertness, some studies have found. This is because darkness is often perceived as dangerous and threatening, causing people to remain alert [52] [53] [54].

Some studies found that brightness, in combination with light colour temperature, can also influence how social you are. In warm bright lighting and cool dark lighting people were more likely to expres other-oriented (prosocial) intentions, while in cool bright lighting and warm dark lighting they showed a preference for self-focused behaviour [55].

The same study found that warm bright and cool dark lighting results in low cognitive depletion whereas bright cool and dark warm lighting increases cognitive depletion. Decreased cognitive depletion leads to greater self-control, increased cognitive depletion diminishes self-control, probably because self-control requires cognitive energy, which is scarce during increased cognitive depletion. [55]

Brightness can also influence personal space requirements. Interpersonal closeness was found to cause significantly less discomfort under high illumination than it did in relative darkness [56]. This suggests that in order for people to feel comfortable in a pub, it should never be too dark.

Perception of the ambience

Low illumination, low CCT and orange accent lighting is perceived as cosier, less lively and less detached than higher illumination, high CCT and cyan/blue accent lighting. [57]

(Wall) paint

Not all pubs are willing to change their interior colour design and, for example, repaint the walls, but some do it regularly. Furthermore, in some cases the same effect could be obtained by using different colours lighting, thus research about the effect of the interior colours is still relevant. Colour design of the interior influences patron’s perception of atmospheric attributes. Customers have a more positive perception of violet interiors than yellow interiors, a study found [58]. another study supports this, saying that short wavelength colours associated with ‘cool’ colours like violet or blue are preferred, which leads to a linear association between affective tone and wavelength [59]. Yet another study has shown that lighter colours are judged as being brighter, friendlier, more cultured, seems to make life easier and more pleasant, and also appear more beautiful [60].

A study found that blue, blue-green, green, red-purple, purple, and purple blue were the most pleasant hues, whereas yellow and green-yellow were the least pleasant. Green-yellow, blue-green, and green were the most arousing, whereas purple-blue and yellow-red were the least arousing [59].

Furniture

When thinking about rearranging the furniture, a pub owner needs to keep fire regulations in mind: fire safety issues often are often cause for limitations in interior design. The type of furniture however can be changed. Pubs often have some bar stools, some have additional seating outside, some have no seating at all. If there is seating outside, it is sometimes accompanied by a table or two. Depending on the pub, tables can find their way into the establishment as well. Then there is the bar, obviously something that cannot be removed. Most of the time it cannot be moved either. Studies have shown that especially the type of seating is important to people: opposing the stereotypical image of pub customers on barstools, sofas are the most preferred seating arrangement inside a venue. Individual seating and bar stools proved far less popular. When asked why, people told that they like relaxing and taking a break from the pushing and shoving every once in a while [42]. Woman like individual seating particularly well, but still less than sofas [61].

Scent

Although pub owners did not think about scent as something to change in their establishment, it can have quite an impact on how people experience their night out. On a night out, one may encounter many scents, both pleasant and vile. Sweating bodies, alcohol, you name it. It would seem logical that ambient scents that mask the vile odours could contribute positively to the night-time experience, and it would not be difficult to implement. A previous study confirms this: they tested the scents of orange, seawater and peppermint, and found that all scents enhanced dancing activity and improved the evaluation of the evening, music and mood [62]. Additionally, they found that the increase in dancing coincided with an increase in temperature in the club. This may lead to the customers wanting more alcohol: if it is hot, what’s better than a cold beer to cool off?

Another study confirmed that adding an ambient fragrance enhances the appreciation and perceived cosiness and decreases the detachment of the ambience. It does not matter which fragrance, the presence of a scent is more important than its nature (as long as it is not a foul scent) [57].

  • Day of the week: heavy drinking occurs mostly on Saturday evenings followed by Friday evenings. This is because young people do not have any responsibilities the next day [63]

Serving style

Patrons drank significantly more beer when purchasing beer in pitchers versus cups or bottles [64]

Factors that influence how much people like a bar

There are some other factors that could possibly be changed in bars, that do not have such dramatic impacts on the behaviour of people but simply influence their happiness and the likeliness that they will enter the bar. There have been a few studies that worked with focus groups in order to find out what people wanted from bars.

First of all we have security. Most bars have security, if only to dismiss all people below legal drinking age. The nature of the security presence outside a bar can be a good indicator of the level of security inside a bar, and the level of security seems quite important. This is because people do not like going somewhere where you do not feel safe. Studies found that a more formal attire offers a greater image of security [42] [65]. People however wanted the security personal to look friendly as well, they did not want to feel threatened by the very security that should protect them [65]. There is a difference here between male and female clientele: while most men belief that more formal attire offers the greatest image of security, only about 50% of woman have the same belief, while the other 50% prefer an informal attire.[61]

The clientele is important as well. People prefer a mix of male and female clientele, with the majority liking a male-dominated place the least [42]. This seems to corollate with the fact that people drink more in mixed-gender groups. Women like a mixed clientele more than men; men like a predominantly female clientele about as much as a mixed clientele [61]. Males offered an explanation for their preference for mixed-gender bars: they identified woman as critical factor in the decision to select a bar [65] [61]. How busy it looked is also named as a critical factor in the decision to select a bar. It should not be too busy, but not to quiet either [61].

The type of venue was found to have impact too. Both men and women prefer a traditional bar, but men prefer this more than women. Women have as close second the wine bar. A Latin themed bar is least liked by both men and women [61] [65].

Preferred location of the dance floor [42] [65]: 1. Surrounded on all sides by people 2. Away from the bar 3. near the bar 4. Surrounded on two sides by people

Conclusions

Knowing all this, a few things are already clear.

There are a bunch of factors that play a part in the overall satisfaction of the patrons, but do not influence anything else. These are things that people like or dislike about a bar, and they mainly determine if they want to enter the pub and if they will have consequent visits. These could be seen as customer requirements for the bar. If too few of those requirements are satisfied, people will be unhappy and could even decide to not enter the bar at all. If a lot of those requirements are satisfied, people will be happy and could even decide to return another day. These requirements are:

  • There has to be background music (background music decreases likeliness to leave).
  • There should be seating (removing seating is bad, people like taking a break every once in a while). If there is seating, sofas provide the most satisfaction, followed by individual seating. Bar stools provide the least satisfaction.
  • Security should be in formal attire (especially male customers find this important, females do not care as long as they feel safe).
  • The clientele should be a mix of male and female.
  • It should be not too quiet and not too busy.
  • There could be a nice scent in the pub (a nice scent in the pub leads to a better evening).


There are also variables that directly influence the alcohol intake of the patrons, or from which the connection to alcohol intake is known. These relations are pretty straight-forward, and one could already make some hypothesis about which value they should have to optimize revenue. These include:

  • Happy hour/price reduction/drink specials increase alcohol consumption. More specifically, 1/2 price leads to 8 times the drinks. Women are more likely to take advantage of drink specials.
  • Louder music leads to a higher level of arousal, which leads to faster drinking
  • Louder music leads to less talking, which leads to more drinking
  • Drinking songs lead to more drinks
  • Faster music leads to faster drinking
  • Patrons drank significantly more beer when purchasing beer in pitchers versus cups or bottles.


Then there are variables that influence internal patron specific variables, like mood, which in turn may influence alcohol intake. These relations are more complicated and delicate, and most variables require further research for their influence to become clear. These include:

  • Tempo of music should correspond to the state of arousal of patrons: people prefer to listen to music that moderates their state of arousal (when in a higher state of arousal, faster music is preferred).
  • More bass leads to more dancing.
  • Colour of light influences mood.
  • Colour of light impacts perception of the area.
  • Arousal (in)congruence of environmental factors influences mood: in case of arousal incongruence, mood decreases. Otherwise, mood increases.
  • Brightness impacts cognitive awareness, which in turn impacts self-control: darkness leads to less self-control.
  • Brightness in combination with light colour temperature impacts how social people are.
  • Brightness influences personal space requirements : darkness leads to people needing more personal space.
  • Colour of the room affects mood:
    • Short wavelength colours are preferred
    • Lighter colours are perceived positively
  • Colour affects level of arousal.


The new variables that emerge are state of arousal, dancing, mood, sociability, and personal space. It is important to find out how those impacted variables further influence the patrons and their bahaviour. In order to complete the picture, in the next section, it will be explained how these variables further fit together.

State of arousal is a variable we already know a few things about. Since it was already established that faster music leads to faster drinking, one might conclude that faster music is desired in a bar setting. But, if the music should moderate the state of arousal of the patrons, it should be ensured that the patrons are in a higher state of arousal. A consequent research question is how to influence the state of arousal of the patrons. The research also shows that arousal congruence of environmental factors increases mood and arousal incongruence decreases mood. This suggests that a colour should be chosen that is of the same level of arousal as fast music, so a colour with a high level of arousal.

Impact of mood on human behaviour

A lot of the factors in a pub that a pub owner can influence, impact the mood of the patrons. This leads to the following question: what is the impact of a person's mood on their behaviour? Previous research showed that mood can influence behaviour via 2 processes [66] [67]:

  • It can have informational effects on behaviour-related judgements and assessments. This in turn results in behavioural adjustment.
  • It can influence behavioural preferences and interests conforming a hedonic motive.

Another study adds that mood influences the style of information processing people use. In a positive, happy mood, people tend to process information in a more heuristic and global manner. This style is paired with a global focus and a reliance on generic, abstract knowledge. Generally, people with a positive mood have a stronger tendency to describe behaviours with afocus on general, superordinate aspects, which are associated with the reasons why to perform an action[68].

Another study suggests that people's interest in behaviours that facilitate hedonic experiences depends on their momentary need for well-being, as well as the perceived effectiveness of potential acts to satsify this need [69] .

In a negative or sad mood, people tend to use a more effortful, careful, systematic, and detail-oriented processing style. This style is paired with a narrowed focus of attention and a shift to a lower level of abstraction. Generally, people with a sad mood have a stronger tendency to describe behaviours with a focus on specific aspects reflecting how to perform it [68].

There as also been studies in the effects of mood on the social behaviour of people, which may be particularly interesting in this context. After all, drinking and clubgoing have a lot of social foundations. Again a distinction is made between happy, positive moods and negative moods.

Research found that a happy person is more likely to initiate conversations with other people, and they express greater liking for individuals whom they have met for the first time. On top of that, people with pleasant mood states are more likely to take risks, on the condition that the risks are not too great and do not endanger their pleasant mood state. This suggests that happy people, who may see dancing as a risk, would be more inclined to dance. Other activities that can be classified as risky, like spinning a prize wheel for a chance to get a delicious drink, would also be more popular among happy people. Other effects of a happy mood appear to be increased creativity, and helping behaviours. The latter suggests that happy people are more inclined to buy alcohol for others [70] [71].

Negative moods however do not have the same consistent effect on prosocial behaviour as positive moods. This is because when in a negative mood, the main motive of people is relieving or disrupting the negative mood states, and their actions will be adjusted accordingly. Since different negative moods need different actions to relieve them, behaviour varies [70] [71].

Conclusions

  • When in a positive mood, people focus on why to perform a task and process information in a more global manner.
  • When in a negative mood, people focus on how to perform a task and pay more attention to the details/
  • In a positive mood, people are more social.
  • In a positive mood, people take more risks.
  • Negative mood can be prosocial but can also be antisocial, depending on which specific mood. Guilt for example leads to prosocial behaviour.

Social structures

Drinking and becoming drunk is a highly sociable activity [72]. As one can imagine, the social context of a drinking occasion can impact (drinking) behaviour [73] [74].

The social structures in a bar mainly concern drinking groups and how people will act in those groups. Some people will go to bars alone, some people will go in large groups. The size of a social group has significant consequences for its dynamics. Conversations become more fragmented as the size of the group increased, and more people drop out of the conversation as group size increased. The proportion of people who are not engaged with a conversation they are physically part of is significantly higher in city centre bars than in community pubs [75]. On top of that, people who are in their local pub or community pubs are in significantly smaller social groups than those who are casual visitors in city centre bars. Those attending their local and those in community pubs are in conversation sized groups (maximum size 4): [76]; [77][78] [79] [80], whereas casual customers and those in city centre bars are typically in parties that were larger than the normative limit for conversations (size > 4). Patrons drink significantly more beer in groups [64].

A larger drinking-group size is associated with heavier drinking [81] [82] [31]. This seems to connect to the previously mentioned conversation dynamics in a larger group: in larger groups, conversations become more fragmented, and more people drop out and decide to drink instead of talk. Another factor that explains the heavier drinking in groups, are rounds of alcohol: with more people, there are more people doing rounds, thus there is a longer time between having to get alcohol yourself while still getting a continuous supply of alcohol. It has previously been found that when purchasing drinks in rounds, especially males tend to consume more alcohol [31][83].

The relationship between the number of friends and the number of drinks is also stronger for men than for women [82], which corrolates with men drinking more when drinks are purchased in rounds.

Gender

On top of the number of friends you take with you, the gender composition of those friends also influences drinking behaviour. A study found that both males and females consume significantly more drinks in mixed-gender groups. [84] [31]. There is a difference between males and females here: men consumed more drinks in groups with an equal amount of males and females and groups with men in the majority. They also consumed more drinks in a group with men only compared to woman only. Woman however, although they did consume more drinks in mixed-gender groups, consumed significantly fewer drinks in groups with men only than groups with woman only [84]. Actually, males were found to drink faster overall. Males drinking beer in bars consumed 0.92 oz per min. Females drank less beer than males, and stayed in a bar for a longer time period [64]. Also, men were found to consume more alcohol than women, particularly at the beginnings of the evening. [82].

There are more differences between male and female pub-goers. One study found that male friends or acquaintances were the main sources of pressure on people to drink or drink more [85]. According to the same paper, pressure to drink also depends on religion and gender. Another study found that indirect pressures to drink play a more significant role than direct pressures [86]. Another study showed that drinkers with companions who consumed large amounts of alcohol tended to consume more alcohol and tended to have higher drinking rates [83] [87].

Females thought it was important to have fun with the group, therefore it is more likely that they stay together and do the activities that most people want to do [88].

Conclusions

  • A larger drinking group size leads to more alcohol. This is a stronger relationship for men than for women.
  • Men consume more alcohol than women, particularly at the beginning of the evening.
  • Men drink faster than women.
  • Both men and women like mixed-gender groups the most to drink in. People hate drinking with only men.
  • Men are the main source of pressure to drink more.
  • Peer pressure happens in the consuming of alcohol, and it leads to people drinking more.

Activities and alcohol consumption

Available activities in bars, like playing games, watching TV or making conversation, has influence on the alcohol intake of people as well. A study found that especially in males, active pastime activities like playing pinball, playing cards or playing table football, result in slower drinking than passive pastime activities like being alone, making conversation or watching TV [31][89]. The males in the study displayed the same drinking rate as women when active, but a faster one when passive. However they compensate for this 'lost time' during the passive activity following the active one by drinking more. They also do drink more alcohol during conversation than females [89].

Why people do certain activities

People dance because[90]:

  • Strongest motivational factor is mood enhancement, followed by self-confidence
  • Women dance for reasons of fitness, mood enhancement, trance, self-confidence and escapism than men.
  • Men were mostly motivated by intimacy.
  • No significant difference regarding socialising and mastery.

Many of the motivational factors related to drinking alcohol also appear in the motivations for dancing, such as mood enhancement, socialising, and escapism. However self-confidence and intimacy are specific to dancing [90].

Most people dance for enjoyment, and prefer dancing with good friends over dancing with someone they are sexually interested in. [91]

Females dance more often than males. Males distance themselves from dancing. This could be because dancing is a very explicit presentation of the self to the public: being in the middle of the dance floor and at the centre of people’s gazes. Sitting in the bar chatting with friends does not involve an explicit involvement in the game of self-presentation. [88]

Active activities and thirst

Active activities in a bar, like dancing, can be seen as exercise. During exercise, sweat is produced, which lowers the amount of bodily fluid. Via a few neurological systems, this results in thirst. The same happens in hotter environments, which could occur due to many people dancing and a limited air conditioning in the summer [92]. Since thirst increases the desire to drink, it is an important factor to consider. While dancing people may not drink that much, but dancing makes thirsty and thirst leads to drinking.

Conclusions

  • During passive activities, patrons consume more alcohol than during active activities.
  • After active activities, males compensate for their "lost time" by drinking slightly faster.
  • Active activities lead to thirst which increases the likeliness to drink.

Alcohol

One thing that will play an important role in the simulation, is alcohol and its effect on people. As is commonly known among many people, drinking alcohol makes one intoxicated, which can have all kinds of influences on people's behaviour [93]. It is important to know which influences to correctly simulate the behaviour of a person over the night as they ingest more and more alcohol. On top of that, each person has a limit to the quantity of alcohol they can ingest before they feel unwell and need to go home, which would be present in a realistic simulation as well.

Alcohol tolerance and the processing of alcohol by the body

The amount of alcohol it takes before one feels drunk, differs per person. It depends among other factors on sex, size, and ethnicity of the person [93]. A study into the number of drinks people needed before they considered themselves drunk found that African Americans reported to need less drinks before they felt drunk relative to whites. Hispanics reported to need more drinks than whites. These effects were stronger in the under 30 group [94]. The same study found that on average, woman reported to need 4.15 drinks to feel drunk. Men needed 6.63 on average [94]. Another study found similar numbers: an average of 5.7 drinks to feel drunk over all drinkers, and an average of 4.4 for women [95]. It follows from this that when simulating drunkenness of people, each person needs to have a unique value for alcohol tolerance.

Intoxication can be observed, but it can also be measured through measuring the BAC, which estimates the degree of intoxication[93]. In figure 1, the association between BAC ranges and the effects of alcohol in the average person are shown.

Fig 1: General association between BAC ranges and the effects of alcohol in the average person [93]

Of course ingested units of alcohol is not a direct indicator for someones BAC, contrary to what fig 1 suggests. For one, as already mentioned, every person reacts differently to alcohol. Secondly, a study found that BAC varies depending on beverage type: it is higher after drinking vodka/tonic than after beer or wine [96]. On top of that, over time the body does its best to process the alcohol. When ingesting alcohol, it of course goes through the stomach, where up to half of it is degraded before the remaining bits are passed into the small intestine. The amount of alcohol degraded in the stomach depends on the level of ADH in the person, which is generally higher in males than in females. Therefore men need more alcohol than women to feel drunk: more ADH means less drunk [97]. A slower rate of absorption can be caused by a strong alcoholic drink on an empty stomach or the presence of fatty foods in the stomach [98]. An alcohol percentage of 20-30% causes the quickest rate of absorption. Also, carbonated alcohol or alcohol in combination with a carbonated drink will be absorbed faster than non-carbonated alcohol [99]. Another factor which can affect the rate of absorption is the mood of the person [97]. The absorbed alcohol is dilluted by body fluids. Therefore, larger people will have a lower BAC after ingesting the same amount than smaller people [97].

Thus, women, small people, and people who drink carbonated alcohol will become drunk faster.

A study found that pre-partying, drinking intentions and number of heavy drinking episodes are significantly associated with patron BrAC. Playing drinking games, patron race, student status, total drinking time, and plans to continue drinking did not seem to be indicators for BrAC [100]. This suggests that a simulation may have to consider people already being a bit intoxicated when arriving at the bar, thus buying fewer drinks before they have reached their limit.


We also have to consider the time span in which the alcohol is ingested. About 15 ml of alcohol is metabolized per hour in an average-sized man [97]. That is about one drink per hour. Thus there is a significant difference between drinking 6 units of alcohol in an hour and drinking 6 units of alcohol in 6 hours: the latter will give you a considerably lower BAC. Thus, if the simulation needs to be close to reality, a time factor should be factored in, where drunkenness goes down when not ingesting alcohol.

Because of its contents, alcohol can serve as nutrient and replenish energy. Per gram of alcohol, about 7.1 calories are released to the body [97]. 6 pints of beer contain about 500 kcal while half a litre of whisky contains 1650 kcal [101].

Alcohol and behaviour

As mentioned before, alcohol can have all kinds of effects on people's behaviour. These effects include:

  • Alcohol produces an effect that may be described as disinhibitory, related to an increase in behaviours that otherwise normally occur at a low rate. Many of these behaviours may be forms of risk-taking that result in aversive consequences to self or others: it was found that choices for the response option defined as risky were systematically increased as a function of alcohol dose [106]. Another study found that expected amount of alcohol consumed had an effect on risk-taking, but actual amount consumed did not. [107].
  • Multiple studies have been done on how alcohol intoxication influences impulsivity. This is often done by investigating degree of discounting, as degree of discounting correlates with tests on impulsivity [108]. While one study found that alcohol had no effect on delay or probability discounting [108], another found that alcohol reduced impulsivity in that no-alcohol participants discounted delayed rewards at higher rates than intoxicated participants [109]. An explanation for this contradiction could be that the first study completed the delay-discounting task five times, which possibly established a stable pattern of responding across the alcohol and placebo sessions, where the second study did not do that. In any case, it is clear that under certain conditions, alcohol intoxication reduces impulsivity.

It was also found that intoxicated participants were more likely to show lack of fit to the hyperbolic model, suggesting they respond less consistently [109].

Why people drink

The psychology behind why people drink could give clues about what factors possibly have influence on the drinking behaviour of people, and is therefore investigated as well.

It was found that drinking gives a valuable opportunity to relax, have fun and form and maintain relationships. As such, primary motivations for drinking include making it easier to socialize, loosen up, or open up [72]. Becoming drunk could also be a way for young people to cope with the stresses and boredom of everyday life[72]. Young people ‘drink to get drunk’. Availability of cheap alcohol and drinks designed to attain drunkenness rapidly (such as shots and shooters) reinforce this norm of ‘drinking to get drunk’ in some contexts [110]. Desires to drink and become intoxicated can also be framed around pleasure or fun[72].

Some people practice predrinking, which is defined as the consuming of alcohol at home (most of the times in groups) before going to a club or bar. This is done because the right level of drunkenness is required to fully enjoy destination bars and clubs [110].

Financial considerations are more important in limiting consumption than concerns about health. Many people restrict how much they drink by putting a monetary limit on the evening rather than one based on alcohol units or an idealised level of drunkenness or sobriety [110]. This suggests that when one wants to increase alcohol sales, they should either get more people to buy alcohol, or they should find a way to convince people to spend more than they planned.

Functions of pub-going

  • Pub-going is associated with stronger social integration in peer networks. Visiting public drinking places is indicative of a way of life which facilitates social integration [111].
  • Public drinking places offer the opportunity to meet the opposite sex and to start a romantic relationship [111]

Conclusions

  • Women feel drunk on average after 4.15 units of alcohol. Men feel drunk on average after 6.63 units of alcohol.
  • About 15 ml of alcohol is metabolized every hour (around one unit).
  • Per gram of alcohol, around 7.1 calories are released to the body. 6 pints of beer contain about 500 kcal.
  • 10 ml of alcohol = 8 gram, so around 56.8 calories are released per hour (apart from the calories in the rest of the beverage, the not alcohol). Around 84 kcal per hours are released per hour including the beverage.
  • Alcohol alters sensitivity to reinforcement and punishment. This is why drink promotions like "spinning the wheel" may work well.
  • Alcohol is disinhibitory, more alcohol can increase dance affinity.
  • Perceived alcohol intake increases risk-taking which can increase likelihood to dance.
  • Under certain circumstances, alcohol intake reduces impulsivity.
  • People drink because they want to relax, have fun and maintain relationships.

Implementation

The full code can be found on Github.

Research variables

Assisted by the research above, the following list of variables was obtained, with their corresponding values and dependencies where present. For each variable, a decision was made to include it in the simulation, or to leave it out. Variables that impact a lot of other variables are often includes, where more independent variables are left out.

General variables

Income

  • Description: The amount of money spent by patrons inside the establishment during the evening/night in order to buy alcohol. Alcohol is the only product available, which has to be bought using euros as currency. At the start of the simulation, this variable will start at a value of 0 euros.
  • Used in simulation: yes
  • Priority: Must have
  • Reasoning: Income is a crucial variable, as the problem statement is based around creating as much income for the establishment as possible. All other variables used during the simulation do relate to the generated income to some extent.


Crowded level

  • Description: Is the establishment filled with patrons or are there barely any people? An upper limit of the amount of people that can comfortably fit inside the bar/dance hall has been estimated (using room size and patron experience). The crowded level is defined as the ratio between the current amount of patrons and the upper limit, displayed as a percentage. At the start of the simulation, this variable will start at a value of 0 percent.
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: As the patrons' happiness can decrease significantly when the bar is overcrowded, it is important to weigh in this factor when evaluating the latter. Patrons can also become less happy when there are barely any other people inside [61].
  • Dependencies: Brightness, number of patrons, room size
  • Impacts: Happiness
  • Value: (number of patrons)/ ((upper limit) + brightness)


Dancing crowd

  • Description: Patrons can start and stop dancing at any time inside the establishment. The Dancing crowd is the amount of patrons that are currently dancing according to their patron state. At the start of the simulation, this variable will start at a value of 0.
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: This variable does not relate heavily to other variables. It is however used when a patron has to decide whether it will start or stop dancing. Because the ability for patrons to dance is an important element of the simulation, the dancing crowd is included.

Environment specific variables

Room size

  • Description: The surface size of the singular room in which all patrons can be found inside the establishment. This room can be used for all patron activities, and is also the location in which alcohol is sold. The specific geometry of the room is neglected. The surface area size is given in cubic meters.
  • Used in simulation: no
  • Priority: Will not have
  • Reasoning: While the size of the room plays a role when analysing an average day at a bar, it is often not the actual size that counts, but the percentage of the establishment that is filled with patrons. For this reason, the room size is accommodated inside the crowded level variable. It is also not something a pub owner can easily change, thus it is of no interest in this simulation.
  • Dependencies: None
  • Impacts: Crowded level
  • Value: As specified by bar owner


Amount of bathrooms

  • Description: A bar or dance hall one or multiple toilets for both male and female patrons. While this is not obligated by the Dutch law to do so (for smaller establishments), an average bar or dance hall will have multiple toilets. This variable does not discriminate between toilets catered to women or men.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: It is determined shown that although people consume less alcohol during bahtroom breaks, bathroom breaks are so short that the "lost time" is easily compensated after the break, thus the total effect of bathroom breaks on alcohol consumption is zero [89]. When a bar or dance hall does not have any toilets, then that might cause customers to not return to that establishment on subsequent nights. Because this research only analyses a singular regular day, it does not affect the simulation results.
  • Value: As specified by bar owner


Brightness

  • Description: A bar owner can adjust the brightness in their establishment. This can impact the patrons in many different ways.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: Brightness can influence different aspects of the patron's state, and could therefore be interesting to implement. However it is not essential.
  • Dependencies: None
  • Impacts: Crowded level, happiness, likeliness to buy alcohol
  • Value: Ranging from -5 to +5, with -5 being completely dark and +5 being completely illuminated.


Light colour

  • Description: Most bars can change the colours of their lights to any colour they desire. Since humans are practically more complicated moths, this can have quite some impact on the patrons.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: The colour of light mainly impacts the mood of the patrons. Since there are many things that impact the mood of the patrons, this can be integrated in the project by determining which mood is most desirable and giving suggestions on how to get patrons to have that mood.
  • Dependencies: None
  • Impacts: Happiness


Paint colour

  • Description: A bar or dance hall can paint their walls in different colours. Studies found that this impacts people's mood.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: The colour of the walls mainly impacts the mood of the patrons. Since there are many things that impact the mood of the patrons, this can be integrated in the project by determining which mood is most desirable and giving suggestions on how to get patrons to have that mood.
  • Dependencies: None
  • Impacts: Happiness


Music volume

  • Description: A bar or dance hall can play music throughout the day. It is assumed that the music genre that is played is appreciated by the patrons. The volume of the music can change, which can alter the behaviour of the patrons. The music can also be absent. This variable describes the volume of the music, which will start at a value of 0 dB.
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: Music volume can greatly change the behaviour of patrons [29][30][31][33], and should thus be included in the simulation.
  • Dependencies: None
  • Impacts: Drinking tempo, likeliness to buy alcohol


Music genre

  • Description: A bar or dance hall can play music throughout the day. Some genres of music, like drinking songs, increase duration of stay and spending [39] [40].
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: The genre of the music being played can alter the behaviour of people drinking at a bar. Mainly the impact of playing drinking songs on the duration of stay and spending in a bar have been found [39] [40], therefore the priority is not very high.
  • Dependencies: None
  • Impacts: Likeliness to buy alcohol, Likeliness to dance, drinking tempo


Scent

  • Description: With the use of an air freshener or some other type of device, it is possible to change the scent in a pub. This has a mostly positive effect on customers
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: While scent does have impact on the patrons, the impact is limited to making the patrons happier. Since the happiness of people is influenced by a lot of other things as well, it is not of high priority to use scent in the simulation.
  • Dependencies: None
  • Impacts: Happiness


Amount of active employees

  • Description: The amount of people currently working in the establishment. While an employee can have different occupations within the bar/dance hall, all of them are generalized under this variable. The employees that are included are all people with occupations that require them to actively engage with the patrons. This includes alcohol vendors situated at the bar and all other employees that are able to assist any patron directly inside the establishment.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: It is assumed that the amount of active employees does not affect the patron behaviour as long as it exceeds a minimum amount, which is dependent on the amount of patrons. Above this limit, all customers can get assistance without having to wait (too long). Below this limit, patrons might become agitated as they might have to wait for long periods of time in order to get the assistance they require. While there is a trade-off between personnel costs and the ability to serve as many customers as possible in a certain time frame, this variable does not have priority as it is assumed

that the owner of the establishment always has enough active employees to exceed this threshold, as to not decrease the patron happiness. As it may have influence on the drinking rate of the customers, it should not be left out completely.


Beverage stock

  • Description: The amount of alcoholic beverages the establishment has in stock ready for sale. Complying with the research constraints, all types of beverages are generalised as one singular type. It is assumed that all patrons are allowed to buy these beverages and do to a certain extent enjoy consuming them. It is also assumed that all products are up to both legal and patron standards, making all of them sellable.
  • Used in simulation: no
  • Priority: Will not have
  • Reasoning: It is assumed that the owner of the establishment has experience and can accurately determine the amount of beverages that are being consumed on an average day. This means that all patrons are able to buy whenever they want. For this reason, this variable does not affect the beverage income and is not included in the simulation.


Amount of seats

  • Description: A bar or dance hall often has seats for patrons to sit on, but almost never enough bar-stools or chairs for everyone. When a patron is tired or does not want to stand, he or she can sit whenever a chair is still available.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: While patrons often try to sit down when they are tired, it does not replenish their energy significantly. People do like relaxing and taking a break from the pushing and shoving, comfortable seating and individual seating are preferred above bar stools [42][61] and adequate seating can increase happiness. However too much seats could be considered obstructions and stand in de way of people who want to get to the bar, which could have influences on the (perceived) crowded level. For this reason, this variable is not a priority, but could be included in the simulation.
  • Dependencies: None
  • Impacts: Happiness


Food availability

  • Description: Is it possible to buy food inside the establishment? Food can range from snacks to real meals, but is generalized as one undefined food item, assuming that all patrons are satisfied with this when requesting food. Food could generate more income besides beverages.
  • Used in simulation: No
  • Priority: Will not have
  • Reasoning: Most bars or dance halls in the Netherlands do not sell food items. Food is ignored in the simulation in order to simulate an average bar or dance hall. Having food as a second source of income would also drastically complicate all variable relations, increasing the scope of the research. Keeping the scope of the research realistic is another reason for omitting food from the simulation.


Specified dance area

  • Description: The establishment can have a specified dancing area. When such an area is available, patrons will not dance on anywhere else. The specific surface geometry or location within the establishment is ignored. The surface area size is given in cubic meters.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: There are no known differences in patron behaviour when a designated dance floor is available. It is therefore unnecessary to implement this variable in the simulation.

Patron specific variables

Money

  • Description: All patrons can carry money, euros specifically. They can then only spend this money on alcoholic beverages. A patron can not spend more money than what they personally have: they can not lend money from other patrons. While real-life patrons sometimes pay for a group, this is not possible in the simulation, as it is assumed to not affect its results.[Buying drinks in rounds does affect amount of beer one consumes,[31][83]] Patrons have a random amount of money on them when entering the establishment, given to them using a [TODO] distribution [mean, var?]. [Can be determined using a survey]
  • Used in simulation: yes
  • Priority: Must have
  • Reasoning: It is essential for the patrons to have money, as it is necessary in order for them to buy alcoholic beverages, thus generating income.


Willingness to pay

  • Description: How easy do the patrons part with their money in exchange for alcohol? Whenever a consumer has less money to spend, he/she is less likely to spend its remaining euros. The value of this variable thus decreases whenever the patron has less money, which might cause him/her to take more time before buying more drinks.
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: While this might have an effect on the amount of money that the patrons will spend throughout the day, this effect would not be large. It is also not proven whether patrons actually decrease their spendings when they have less money. They will likely be under influence when this happens which might decrease their good judgement, and actually increase their willingness to pay.
  • Dependencies: Intoxication
  • Impacts: Likeliness to buy alcohol


Has alcohol

  • Description: Describes whether the patron currently has alcohol on his/her person. A patron will not buy any more alcohol whenever he/she already carries alcohol.[friends sometimes buy alcohol for eachother, which can result in "2 in de hand"] When carrying alcohol, the patron might consume parts of it, which slowly intoxicates him/her. Whenever the alcohol is completely consumed, the patron has no alcohol anymore.
  • Used in simulation: yes
  • Priority: Must have
  • Reasoning: In order to generate income for the establishment owner, it is important that the patrons can buy beverages. It would be unfair if they could buy drinks and not consume it afterwards. Furthermore, this better mimics reality, as real life patrons generally consume their products as well.
  • Dependencies: Drinking tempo, time
  • Impacts: Likeliness to buy alcohol
  • Value: The variable is defined as a percentage, with 100 percent denoting a full beverage, and 0 percent the absent of any drink.


Intoxication

  • Description: A patron can become intoxicated when he/she consumes alcohol [94][93]. This variable is represented as a percentage, with 0 percent representing a completely sober individual. When the patron consumes enough alcohol and reaches 100 percent, he/she will leave the establishment immediately. Intoxication changes the behaviour of a patron [93].
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: Intoxication alters the behaviour of the patron significantly. It also might cause a patron to leave early whenever they become too intoxicated. For these reasons, this variable is implemented in the simulation
  • Dependencies: Alcohol tolerance, drinks consumed, time
  • Impacts: Likeliness to dance, likeliness to buy alcohol
  • Value: ((Drinks consumed) - (1 per hour))/(alcohol tolerance)


Alcohol tolerance

  • Description: This will show how much a certain patron can tolerate drinking alcohol. This is different per type of patron, some people can better stand drinking alcohol whilst not completely going drunk [93][94].
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: Everyone has a different alcohol tolerance, to determine when Intoxication reaches 100% we need to know this tolerance.
  • Dependencies: Gender
  • Impacts: Intoxication
  • Value: Average 5.1 for women, average 7.4 for men


Drinking tempo

  • Description: This will show how fast a person is drinking. Different factors have impact on how fast a person drinks. The faster a person drinks, the faster he is ready for another beer.
  • Used in simulation: Yes
  • Priority: Could have
  • Reasoning: Do determine when a patron finishes a drink it is needed to know how fast that patron drinks.
  • Dependencies: Music genre, music volume, gender, state
  • Impacts: Has alcohol
  • Value: 0.92 oz/min for males, slower for females.


Group sizes

  • Description: This will represent with how big of a group you are there. Since it will be more enjoyable to a patron, when the patron can converse and dance with more people he knows [72].
  • Used in simulation: no
  • Priority: Could have
  • Reasoning: At this moment it will not be included, however when the simulation would allow it this could be included as another variable.
  • Dependencies: None
  • Impacts: Likeliness to buy alcohol


Happiness

  • Description: Happiness will depict how much a patron is enjoying himself. This can changes because of other variables in the simulation, for example when dancing [91] and talking happiness will increase. This can be represented as a percentage in each patron, which will have certain threshold for leaving or staying.
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: With this it can be modeled how people are enjoying themselves. Which is an important part of the simulation. Since with this can be determined if people will stay when they are happy or go when they are unhappy.
  • Dependencies: Brightness, light colour, paint colour, state, intoxication, scent, crowded level
  • Impacts: Likeliness to leave, likeliness to dance
  • Value: happiness = (0.5 * intoxication) + (0.5 * energy), when dancing happiness has a base value of 65%, the other 35% is determined by previous formula.


Dance affinity

  • Description: Dance affinity is here to show how much a certain patron enjoys to dance. Certain people will have a higher tendency to dance, whilst others will only dance in the right situations and circumstances.
  • Used in simulation: simplified
  • Priority: Should have
  • Reasoning: This variable will be relatively small and would prove to complex to model currently. Also since the variable is likely to change in the simulation for a patron, due to for example drinking alcohol and thus becoming more loose. This will however be represented by a the same constant value in each patron.
  • Dependencies: None
  • Impacts: Likeliness to dance


Thirst

  • Description: A lot of the time, people drink because they are thirsty, and want to quench their thirst. People can get thirsty for a lot of reasons, and being thirsty can be a compelling argument to buy another beer.
  • Used in simulation: no
  • Priority: Should have
  • Reasoning: This variable could be used in the simulation, since it would be a great way to describe when a people stops dancing to get beer.
  • Dependencies: State, time, has alcohol
  • Impacts: Likeliness to buy alcohol
  • Value: Relaxing: thirst = thirst + (0.5 * time); Dancing: thirst = thirst + (0.7 * time); thirst = 0 if the patron has a drink;


Energy

  • Description: A Patron will have energy, with which they can do actions. When they do an actions, energy is subtracted from the energy of that patron. Possible action for example are dancing, standing, talking, drinking and buying. Which all have different energy costs. This will likely be used as a certain value from which patrons can differ from, with a small difference from the mean.
  • Used in simulation: yes
  • Priority: Should have
  • Reasoning: This variable is needed in the simulation, since people who are tired are likely to go home
  • Dependencies: State, time, intoxication, drinks consumed
  • Impacts: Likeliness to dance, likeliness to leave

Patron states

Five states have been defined for the patrons:

  • Drinking: The only state that can be done simultaneously with any of the other states: In this state, the patron's alchohol will be drunk, increasing its intoxication level.
  • Dancing: The patron is currently dancing and socialising with others. This actions does cost more energy than any other action.
  • Talking: The patron is conversing with others, This does cost some energy.
  • Buying: The patron is currently buying more alcohol for him/herself.
  • Standing: In this state, the patron is inactive and unsocial. This state barely costs any energy.
  • Outside: The patron is outside the bar.

At all times, each patron in the simulated bar will be in exactly one of these states. They will start in the state "outside". This state is added to the simulation to keep track of the people who are not in the bar. As such, people who have left the bar, will still exist with all their inner variables intact. Thus, would they decide to enter the bar again, they are the same person, instead of an entirely new one. Upon entering the bar, the agents will be in the sitting/standing state, in which they will talk with each other and drink their beer, if they have one. Each step, the likeliness to go to another state will be checked. When this likeliness reaches a certain value, they will move from the sitting/standing state to the corresponding state, and will perform the actions specified for this state. In the dancing state, they will move to the dance floor and "dance", in the buying state, they will move to the bar and buy a beer, and in the leaving state, they will exit the bar. As all agents have different values for their internal variables, agents will display their own behaviour when moving from state to state. The state diagram in figure 2 shows the relationship and transitions between those states.

Fig2: state diagram displaying agent's states and transitions

As can be seen, all transitions have conditions. If the condition is satisfied, an agent will transition from one state to the other. These conditions are defined by four "variables":

  • Likeliness to buy alcohol
  • Likeliness to dance
  • Likeliness to leave
  • Likeliness to relax

These, in turn, can be described using the previously obtained variables as follows:

Likeliness to buy alcohol

Likeliness to buy alcohol depends on a few things.

  • First of all, the variable "has alcohol" should be close to 0: people generally do not buy more alcohol when they still have alcohol.
  • When brightness increases, likeliness to buy alcohol decreases
  • When music volume increases, likeliness to buy alcohol increases
  • When a drinking song is played, likeliness to buy alcohol increases
  • Lower alcohol price/drink deals increases likeliness to buy alcohol.(1/2 price is approximately 8 * alcohol) This is stronger for women than for men.
  • When other people in a person's group have alcohol, likeliness to buy alcohol increases. This is stronger for men than for women.
  • Males are more likely to buy alcohol than females.
  • When thirst increases, likeliness to buy alcohol increases.

Likeliness to buy alcohol is a value between 0 and 100. When first entering the pub, likeliness to drink is on average 60, since most people entering a club want alcohol. It will then be increased or decreased based on environmental factors and actions of the patron.

Likeliness to dance

Likeliness to dance depends on a few things:

  • If more people in the group are dancing, likeliness to dance increases
  • An intoxication level between tipsy and drunk increases likeliness to dance
  • An intoxication level nearing "af" decreases likeliness to dance (increases likeliness to vomit, cry, and generally have a bad time)
  • A lower energy level decreases the likeliness to dance
  • If a dancing song is played, likeliness to dance increases
  • Higher levels of happiness increase likeliness to dance
  • A higher dance affinity increases the likeliness to dance
  • Women are more likely to dance than men.

When entering the pub, the likeliness to dance will be equal to the dancy affinity of the patron.

Likeliness to leave

  • Lower levels of happiness increase likeliness to leave
  • Lower levels of energy increase likeliness to leave
  • Higher levels of intoxication increase likeliness to leave
  • If a high number of friends is in the state "outside", likeliness to leave increases
  • If music volume = 0, likeliness to leave increases

Likeliness to relax

  • Low energy increases likeliness to relax
  • A high number of friends in the state "standing/sitting" increases likeliness to relax
  • A high value of "has alcohol" increases likeliness to relax
  • Slower music increases likeliness to relax

User Interface

The simulation will be offered to users (bar/club owners) as a consultation service. This decision drastically affects the requirements and structure of the user interface. The purpose of the inclusion of the consultation service is to shift difficult and/or time consuming procedures away from the user in an effort to create more appeal to the product. The user will now only have to deliver specifications about their establishment and expected customers as input. The entire system can then be treated as a black box for this user, from which the consultancy will extract and send back output that can help the user to improve their generated income.

Inputs

The user will receive a form with a list of variables that the consultancy requires of him/her in order to properly simulate the establishment environment. While most of these values will be required, some are optional. All variables on the form have altered names have accompanying descriptions and expected ranges or values if possible. This could help the user to better understand the requirements and give more accurate values. The required and optional inputs, complete with description and expected ranges are as follows:

Required

  • Establishment opening and closing time
    • Description: Give the range of hours at which the establishment is open to patrons. Times only required for one day: if bar times vary a lot per day, choose an average
    • Expected value: 18:00 - 4:00
    • Translates to: Simulation starting and closing time.
  • Establishment size and shape + door position
    • Description: Give information about all wall segments of the bar room: their length and angle. Also include the position of the entrance/exit inside the establishment
    • Expected value: establishment sketch with parameters
    • Translates to: Wall parameters, patron spawn and leave location
  • Bar / Alcohol vending size and location
    • Description: Describe where inside the establishment with respect to the walls the bar is located. The size, rotation and location are necessary parameters
    • Expected value: establishment sketch with parameters OR X by Y meters, positioned at Z
    • Translates to: Bar table parameters
  • Dance area size and location
    • Description: The floor area on which patrons can dance. This can be a real dance floor or an (open) area at which the owner expects them to dance. This area can not be the entire floor area, as patrons require room to rest and converse without dancing.
    • Expected value: establishment sketch with parameters OR X by Y meters, positioned at Z
    • Translates to: Dance area parameters
  • Other objects / walls
    • Description: Other objects that have to be paid attention to when walking through the establishment. These objects will have to be walked around.
    • Expected value: establishment sketch with parameters OR X by Y meters, positioned at Z
    • Translates to: Inclusion of extra collision objects with described parameters
  • Beer vending price
    • Description: The price of a regular tap beer.
    • Expected value: 0 to 5 euro
    • Translates to: Alcohol unit cost
  • Music volume
    • Description: The average volume of the music inside the establishment throughout the day/night.
    • Expected value: 0 dB to 100 dB
    • Translates to: Music value from 0 to 1

Optional

  • Number of bar stools
    • Description: The amount of bar stools present at the bar
    • Expected value: 0 to 20
    • Translates to: The amount of bar stool objects inside the simulation environment
  • Energetic clients?
    • Description: Are the clients on average energetic: do they dance and walk a lot?
    • Expected value: No, A little bit, Yes
    • Translates to: Average patron energy percentage spending increases or decreases
  • Patron average spendings
    • Description: An estimate of the average spending amount of a patron on a single day/night
    • Expected value: Between 2 and 200 euros
    • Translates to: Altered patron money when entering to reflect the spending amount
  • Peak hours
    • Description: The hours at which the establishment has the most customers. User is allowed to further describe at what hours people usually enter/leave the establishment
    • Expected value: 21:00 to 1:00. People start entering around 17:00
    • Translates to: A patron arrival scheme which results in the most present patrons during these hours

The received input is then translated by an expert working for the consultancy service to variable values inside the simulation. The consultancy interface makes sure that there is no need for a simplification of the simulation. As the expert is familiar with handling the input and reading the output inside the program, no simplified UI or internal translation from 'human' variables values to simulation variables is required.

Outputs

The outputs are gathered from the information provided from the simulation. These variables are extracted form java and than plotted in a figure. From these figures can be seen how an output changes through the course of a night. There are also optional output for the client that could be helpful in providing some insight. The main and optional outputs are presented below with their description:

Main

  • Total money spend
    • Description: This shows how much money is being spend by the customers in one night, therefore this is also the amount of money earned by the pub owner as revenue.
  • amount of people
    • Description: With this the pub owner can see how many people there are during the night in the bar.
  • amount of people dancing
    • Description: This shows how many people are dancing through the night, which can be used as an indicator to how much fun people are having.

Optional outputs for every customer

  • Energy
    • Description: Energy is used to show how much energy a person has, this variable changes in the course of the night due to actions a person makes. Decision of persons in the simulation can change on how much energy they have, therefore this could be an interesting output for the client.
    • Expected value: Energy will be normalized, therefore will range from 0 to 1.
  • Position
    • Description: This will show of each person, where during the night the person is.
    • Expected value: Depends on the size of the pub.
    • Translates to: Pub parameters, patron spawn and leave location
  • Happiness
    • Description: Happiness shows how much a person is enjoying themselves. Based on the happiness of a person different decisions will be made, if a person becomes to unhappy they will leave the pub.
    • Expected value: This will be normalized, therefore will range from 0 to 1.

Processed results

From these output variables figures are made, depicting how these values change over the course of one night. These can give the barowner insight in how their bar operates and because the barowner already knows a lot about their own bar it will give the barowner an understanding of how accurate the simulation is.

Then we can change up variables within the range of possiblities of the barowner, giving us a new set of output variables. From this the impact that the inputs have on the outputs can be determined, by comparing the different figures with each other. By doing this it can than be determined what changes would be more beneficial to the client. As an consultancy we would also be able to give additional tips to the client, which where learned from previous researches and cases. Meaning that by comparing the situation to previous clients, we can find out if any had similar situation and based on that provide additional advice. Therefore based on experience and the findings of the comparison, ways of improving the pub can be presented to the client.

Experiments

Three types of experiments are conducted to verify the correctness of the simulation and to analyze the possibilities of the simulation.

1. Verification of the simulation

One of the requirements for the simulation is that it should accurately reflect real life. This is quite important: if the simulation does not accurately reflect real life, generated outcomes may not be useful for the user at all. Tests will be performed to verify the accuracy of the simulation, as well as the consistency and the sensitivity, as described below.

1.1 Consistency test

To verify the consistency of the simulation, the simulation will be run a few times without changing any input values. The output from all different runs will be gathered and compared to each other. A mean value as well as a standard deviation will be calculated. This allows for the opportunity to analyze the stability of the output. If different runs with the same inputs generate wildly different output values, something is wrong, since no output is generated randomly and should therefore not be random. At the same time, it can be checked if certain output values are in expected ranges or if they are in extreme ranges. If, for example, the mean total money spend by patrons is zero, something is clearly wrong with the way money is processed. This finding can then be used to improve the simulation.

1.2 Sensitivity analysis

To verify that there are no variables that have an unreasonably big impact on the outputs, a sensitivity analysis is conducted. The simulation will be run a couple of times, and in each run a different input variable will be adjusted to have a slightly different value. If one of these adjustments lead to a profoundly different value for one or multiple output variables, this could mean something is wrong. In this case, further analysis is required to check if the input variable in question should have such an impact on the output variable in question, or if the simulation is inaccurate.

1.3 Verification of patron specific variables

The exact value of some patron specific variables could not be found in the results of previous studies. Thus it had to be based on educated guesses. To verify the correctness of these guesses, the simulation will be run a few times. In each run, one of the estimated variables is adjusted. The effects of these adjustments on the output variables is then analyzed. Based on this analysis, either a new value for the variable is found or it can be confirmed that the variable had the correct value.

2. Insight into current situation

Once it is established that the simulation is correct, more elaborate tests can be done to analyse and explore the functionality of the simulation. To get insight into the current situation for an average bar, the simulation will be run with as input parameters a "normal" pub environment. The values for a "normal" pub environment are based on personal experience and the research shown above. It is important to establish what output values are "normal" for an average pub. After all, when it is unknown what output values are "normal", it is impossible to decide what is optimal. The changes in the output values "total money spend", "average happiness", and "alcohol consumption" over time will be studied and analysed.

3. Optimizing the bar environment

Once it is established what is "normal" for a bar environment, an analysis will be made about which impact certain input values have on the total money spend and the happiness of the patrons. A few input values are selected to analyse: opening hours, size of the bar, number of people in the bar, and music volume. The impact of a variable is analysed as follows:

  • All variables in the simulation are reset to their default, "normal" bar values, as defined in the code.
  • The variable in question is changed to its desired value
  • The simulation is run a few times, after each time the output variables "total money spend" and "total happiness" are gathered
  • The average of "total money spend" and "total happiness" over all runs is calculated. These values are plotted in their corresponding graphs.
  • The process is repeated until all desired values of the variable in question are tested

The obtained graphs will make it easy to see for which value the total money spend and total happiness are optimal.

Results

test

  • Description An example of a simulated bar has been created, with estimated values for both the environment and patrons. This bar was then simulated multiple times, in order to test its consistency.
  • Consistency result 1, 2 and 3 originate from simulation runs with identical input variables. The mean amount of generated income came down to 686 euros. The standard deviation was then 33 or 4.8 percent.

From multiple identical simulations, the average deviation of the mean

  • Additional observations While input values were identical for each run, the virtual patrons that are generated with these values can still differ per run. This creates a slight inconsistency, which has to be taken into account when analysing the other results.

test

  • Description An example of a simulated bar has been created, with estimated values for both the environment and patrons. This bar was then simulated multiple times, in order to test its consistency.
  • Consistency result 1, 2 and 3 originate from identical simulation runs. The mean amount of drinks consumed came down to 626 drinks. The standard deviation was then 63 or 10.2 percent.
  • Additional observations From previous figure can also be seen that these 2 figures look alike, this is due to alcohol costing 1 euro. The small difference are due to people not 'chugging' their alcohol.

test

  • Description Shows amount of people in pub in results 1,2 and 3.
  • Consistency All input variables are identical in each run. These results show that, while peak hours can still be recognized, exit times of each patron are not that consistent.
  • Additional observations The inconsistency between the results could again originate in the fact that generated patrons have randomized values and can therefore vary even when the input values are identical.

test

  • Description Shows amount of people dancing in results 1,2 and 3.
  • Sensitivity While no input variables have been altered, something can still be said about the sensitivity of the number of people dancing with regard to the total crowd (previous graph). It is interesting to see that the three results of the amount of people dancing do consistently follow the amount of patron results up until the final hours, in which the high peak of results 2 and 3 are not reflected when looking at who is currently dancing.
  • Additional observations These results are heavily tied to the amount of people that are currently inside the establishment. This connection is not directly programmed in but a result of multiple variables and smaller systems working together.

test

  • Description Shows the energy of one person from result1 in previous figures
  • Additional observations From this graph, it becomes clear that the built-in patron states are functional but rudimentary. The change of behaviour is in some states almost completely based on their energy level. This causes the patron energy to slowly switch between two values as the patron goes back and forth between two states. It can be seen that there are still other factors playing a role, as the energy does not always follow this pattern.

test

  • Description Shows the happiness of one person from result1 in previous figures
  • Additional observations The happiness of the patron is influenced by multiple variables and better integrated into multiple states than the energy variable described above. It can also be seen in the results that the patron happiness over time behaves a bit more complex.
Top view of the position, to show the places where the person goes to.

test

  • Description Shows the position of one person from result1 in previous figures
  • Additional observations The vertical line at the beginning of the day shows that this patron did not enter the establishment until roughly 3 hours in. After this period of time, The patron had 3 areas at which he preferred to go to: the dance floor, the bar and the conversation area. These areas with connecting paths between them can be seen in the results.

test

  • Description Different types of pubs are tested to show what a difference, it might make. the standard bar is 60 by 160. The bigger bar is 40 by 40 'units' bigger, with normal dancefloor in pub. the bigger dancefloor is 50 by 50 'units' bigger (originally 200 by 150), with normal bar in pub.
  • Sensitivity A bigger bar, which was increased in volume by roughly 20 percent, resulted in a change of income of 42. A bigger dance floor, which was increased in volume by roughly 30 percent, resulted in a decrease of 56.
  • Additional observations The extra volume of the bar was added on the bar-side close to the dance floor. The extra volume of the dance floor was added on the side that was removed from the bar. It might be that the volume does not matter as much as the location of the additional volume, as patrons have to on average walk a longer/shorter distance between the points of interest.

test

  • Description With this simulations the amount of people entering was set to different values, to show what a difference in customers would have on the revenue generated.
  • Sensitivity It seems that the sensitivity is higher when the number of patrons is low, and decreases to an almost constant value between the 100 and 300 people.
  • Additional observations The high sensitivity at low crowd sizes is especially important for small establishment owners.

test

  • Description With this simulations the amount of money people have was set to different values, as shown in figure, to show what a difference it would make on the revenue generated.
  • Sensitivity It seems that the sensitivity is higher when the money range values are low.
  • Additional observations Attracting only patrons with little intention of spending money is not advised.

test

  • Description With this simulations the music volume was set to different values, to show what a difference it would have on the revenue generated.
  • Sensitivity Even with the non-linear mapping between the music input and the income output, it is clear that the sensitivity is higher for a low music volume than for higher volumes.
  • Additional observations The results seem to suggest that there is a volume at which the income will peak. That volume in this example simulation seems to be close to 75dB.

test

  • Description With this simulations the speed with which people drink their alcohol was set to different values. For all these simulation the slowest they will do is 7 sips an hour, the max for these 3 are different. So for slow drinkers the max they can drink is one sip every two minutes. For normal drinkers the max is every minute. For fast drinkers the max is every 30 seconds. To show what a difference in drinking speeds would have on the revenue generated.
  • Sensitivity Doubling the drinking speed seems to roughly double the generated income. While the results seem to suggest that the sensitivity decreases as the drinking speed increases, this can not be confidently concluded as there are too little results.
  • Additional observations This result is not not unexpected. Nonetheless it shows that this variable in isolation has a high chance to accurately mimic the effects of its real-life counter parts.

Conclusion

1. Verification of the simulation

1.1 Consistency test

The consistency tests showed that most outputs were quite consistent. Any inconsistencies are likely caused by the fact that the generated agents are all different, which leads to different actions and, consequently, different outputs.

The amount of people was less consistent. In each run, the amount of people in the bar followed an entirely different pattern. This could be explained by the aforementioned differences between patrons in different runs: patrons with certain attributes may have been more inclined to leave. This inconsistency is therefore not one to worry about: it does not necessarily mean that the simulation is faulty. It may even confirm the accuracy of the simulation. It does however show that when analysing the simulation, it should be kept in mind that differences in output will not always be caused by the tested input variable: it may just be caused by the inherently different agents. From this, it can be concluded that the consistency of the simulation is correct.

1.2 Sensitivity analysis

The sensitivity test showed that there are some variables that are sensitive to other variables.

The amount of people dancing corresponds to the number of people in the establishment: when there are more people in the establishment, more people are dancing. In the beginning of the evening, relatively more people are dancing than at the end of the evening. This shows that the amount of people dancing is quite sensitive to the amount of people in the establishment, which is quite normal.

It was interesting to see how when increasing the amount of people from 50 to 100, the total money spend was tripled, but when increasing the amount of people even further this trend was not continued. Increasing the amount of people from 100 to 200 only roughly doubled the amount of money spend.

A similar thing happened when increasing the amount of money people have. The first, smallest, increment increased the total money spend with more than 100, while subsequent bigger incrementations only increased the total money spend with 50. An explanation for this could be that in the first range, people could have 0 money, and therefore not be able to buy beer. Having non-drinking people in your establishment sure is not good for business. When all people have at least 5 euros, all people are able to buy at least 5 beers, which heavily increases sales.

Total money spend was also quite sensitive to the volume, but only when going from a low to a medium volume. Again this can be explained by the amount of money people have: people may be buying more beer when the volume is higher, but when they have spent all their money, they cannot buy beer anymore. This might happen more often with a higher music volume. Since there are no variables that were found to have an unexpectedly profound impact on the output variables, one can conclude that the simulation is accurately sensitive.

1.3 Verification of patron specific variables

The tests to verify the patron specific variables showed that the correct variables were chosen, as others gave more unlikely output values.


2. Insight into current situation

The tests to get insight into the current situation showed that the mean amount of money spend in the “normal” establishment is 686 euro, with a standard deviation of 33. The mean amount of consumed drinks was 626, with a standard deviation of 63. The amount of people varies greatly, as does the amount of people dancing.


3. Optimizing the bar environment

The tests to optimize the bar environment mainly showed that more is better. Increasing input variables leads almost all the time to an increase of total money spend. A bigger dance floor however decreases the total money spend, which could be due to the fact that dancing agents do not consume alcohol. It could also be due to the fact that agents had to walk further, which leads to more time before an agent has alcohol. Louder music does not always increase the total money spend either. Up to a volume of 75 dB, the total money spent increased when increasing the volume, but after that it decreased. A music volume of 100 dB lead to less money spent than a volume of 50 dB.

Discussion

General notes

Although contact with people knowledgeable of the inner workings of a pub was reached, due to coronavirus it was difficult to actually visit a pub or perform an observational study in a pub. This limited the confirmation of our findings to literature studies and the input of those people knowledgeable of the inner workings of a pub. For future research, observational studies in pubs could be carried out to confirm the accuracy of the implemented simulation and to adjust variables. This could lead to the expansion of the simulation to resemble real life more closely, since some variables had to be estimated due to limited available research

Establishment improvements

Various simulation aspects connected to the establishment are currently ignored or could be better defined. Most real world bar or dance club strictly sell alcohol at the bar table. In order to prepare the products and to exchange it for money with the patrons, The bar table requires beer taps, employees and sufficient stock. The effects of these components on the rest of the simulation have been estimated and integrated in other aspects of the simulation. However, to improve the accuracy and reliability of the simulation, these limitations need to be addressed. In the future, these variables could be made more complicated and more delicate interactions between the environment and the patrons could be implemented.

Alcohol taps

The amount of beer taps present in the establishment determines how many patrons can order a drink at the same time. To prepare a drink using the beer tap also takes time. During this time, no other patrons can order any drinks at this tap. When the establishment is relatively empty, the total time that anyone would have to wait for their drink is not that long. But this total would grow almost exponentially when the crowd grows. As the time to tap a beer is also influenced by the speed of the bartender, and because the amount of employees also influences how many taps can be operated at one time, the beer tapping has been combined with the employees into one system inside the simulation. These have been modeled as an expected waiting time, given the amount of current patrons. This model is inaccurate however when compared to the real-life situation. An improvement would be to model these taps and their tap time separately from the employees.

Employees

The extent to which bartenders were simulated, was very limited as well. While in real life, bartenders are real people with or without charisma which bring all kinds of unknowns into the equation, in this simulation they were merely implemented as a waiting time. This was too avoid making the model too complicated, but it might have affected the results of the simulation. In the future, it could be investigated to which extend bartenders can influence the choices of the patrons, and what the effect of hiring charismatic people is.

Product stock

Because of the time limit of the project, a few decisions to simplify variables were made. Instead of serving all kinds of beverages a bar usually serves, the proposed simulation was limited to beer. This may have had some effect on the behaviour of the patrons: as was found in the research about alcohol, the processing of beverages with a higher alcohol percentage is different from the processing of beer. Not offering the possibility to "sober up" using a non-alcoholic beverage may have affected the results as well. In the future, the proposed simulation could be extended with different choices of beverages, and the effect of serving non-alcoholic beverages and shots could be investigated. The exclusion of an actual product stock inside the simulation is justified according to our advisor (bar-owner). In his experience, a bar should always have enough stock for at least the current day and should therefore almost never run out of stock. As the simulation tries to mimic a real life bar as accurate as possible, alcohol shortage would never occur and is therefore unnecessary to include. Ignoring this aspect completely then helps to keep the simulation as simple as possible, which helps both the consultant experts and the client to better understand what aspects have impact on the generated income.

Future model extensions

1. Extended interaction between patrons Currently, patrons decide on how to act almost exclusively based on the bar environment. An important extension to accurately simulate a real life bar crowd would be to improve grouping and in-group behaviour of these patrons. Patrons should be able to arrive in groups, or arrive separately but still belong to a group. When groups are included inside the simulation, different inter-patron behaviour can emerge from this that mimic real-life counterparts. One example is that people in groups often position themselves close to each other, usually in circles, and perform the same type of actions. In the simulation, patrons can look at what their group members are currently doing and use that in their evaluation of what state they should be in. If members are dancing in the middle of the dance floor, then other members more likely to join in.

Another example of in-group behaviour would be to let one group member buy a round of alcohol for the other members. If only one member has to go to the bar table and buy alcohol, than other members dont have to potentially stand in line, increasing their happiness. Members might also peer pressure each other into drinking the whole beverage at once, which could improve the alcohol vending. Other in-group behaviour has also been observed according to multiple studies, as in the research shown above (83-88). These conclude that grouped patrons consume their beverages faster and keep drinking for longer than individual patrons. This would certainly have an effect on the generated income and would thus be valuable to include in the simulation.

Possible extensions that can further improve accuracy but do not contribute as much as above are the following. Patrons that belong to one group could still fraternize with people from other groups. It might even be possible to be part of multiple groups in which case the patron can move between the two groups during the evening/night. Another small improvement would be to let patrons have other emotions for each other than to just be friends or indifferent. This could result in small fights between two or more patrons. It could also result in flirting, which undeniably happens in real life bars as well.

2. Behaviour extension of the patrons given the bar environment Besides interacting immediately with each other, patrons could also be influenced by the bar environment and the crowd in its entirety. This can be better fleshed out with multiple elements. For one, the bar stools do not affect the patrons. Patrons that are tired or waiting at the bar can sit on the stools in order to regain energy. The amount of stools would then be a factor that the user might be able to change in order to increase the income. Another element would be the ambiance of the establishment. This could tie in with the music and the emotions of the patrons. Happiness can be regained if certain music genres are played. This could entice the patrons to drink faster or dance more.

3. Bar environment alterations given the patron behaviour Currently, many aspects of the environment are static. While most of these should remain, such as wall sizes and positions and object positions, some aspects would be more realistic if they become dynamic. The establishment is for example always clean, while is would certainly become dirtier throughout the day in real life. Patrons are however not often disturbed by the mess and do therefore not get affected that much. For this reason, this aspect is not as important to model inside the simulator as many of the other extensions mentioned in this section.

4. Inclusion of other establishments Currently, the establishment exists isolated from the rest of the world. The regular amount of people will join at roughly expected times, and leave later through the day/night. In real life, every bar or club has to compete with other establishments over customers. Having more customers generally means more income, which is desirable for the client. This gives reason to the inclusion of a simulated group of establishments. This does however take a lot of effort and research in order to get accurate results. For one, it should be determined what aspects of the establishments are important for the patrons when deciding where to go. These aspects should then be used in order to simulate believable other bars/clubs. The group of all patrons should then be extended to a crowd that can decide when and where to go. This requires that all patrons should be able to leave one bar and enter another later through the day/night. One final aspect that would be interesting to add are regulars. These patrons will always choose the same establishment.

5. Extension of patron behaviour complexity The patrons mental and physical state is currently mostly represented through their state and happiness. Real life patrons are observed to have a more complex behaviour than currently simulated. By including more emotions, and increasing the amount of possible states that the patron can have, more complex behaviour should be able to be simulated. Real life patrons can for example flirt with other patrons. These patrons would try to stay close to each other and possible dance/talk together. The gender of each patron could even play a role: research data could be use to determine what percentage of patrons could flirt with any gender. It might also be possible to allow patrons to dislike each other. This could cause them to avoid each other or confront them, decreasing their happiness and of the patrons standing close. While it can not be said with certainty what effect these aspects would have on the generated income, it might be interesting to conduct research on this topic and it creates a more realistic simulation.

Appendix

Planning

  • Week 1: research state-of-the-art, finalise plan
  • Week 2: More research, state requirements for simulation, create research plan for filling in the blanks of the state-of-the-art.
  • Week 3: Implemented first version of simulation with only basic features
  • Week 4: Implemented second version simulation with requirements implemented
  • Week 5: Performing simulation, documenting results
  • Week 6: Performing simulation, documenting results
  • Week 7: Compare results to real life, create conclusion
  • Week 8: Finalise wiki


Gantt chart

Milestones

Tasks Estimated Time
Planning 12:00
State-Of-The-Art research 75:00
Conclusions state of the art 5:00
Set up research plan 5:00
Execute research plan 50:00
Analyze results 20:00
Minimum viable product 50:00
Simulation with more complex AI 100:00
Analyze simulation 75:00
Update research simulation based on results 50:00
Write down conlcusions 20:00
Finalize wiki 20:00
Create presentation 10:00
Unforseen 100:00
Total 592:00

Deliverables

State-of-the-art

A list of variables influencing behaviour at a bar setting.

A list of missing variables influencing behaviour at a bar setting.

A hypothesis on how the results our simulation would create.

A research plan detailing how to fill in the blanks of the state of the art.

Results of our research.

A minimum viable product: a simulation that houses AI that can navigate a bar setting.

Extension of the minimum viable product trough implementing more complex behaviours found in the research.

Results out of the simulation, a set of factors and the behaviour it creates in the simulation.

Finalized wiki and conclusions.

A presentation


The latest version of the model can be found at https://github.com/JasperDell/Robots-everywhere

The presenation can be found at https://youtu.be/ijEgbMvziV4

Who is doing what

Week 1

Name Total Break-down
Daan 2h Discussing the subject
Job 2h Discussing the subject
Sanne 5h Making a draft for the wiki (0.5 h), Gathering links for the State of the Art (2.5h), Discussing the subject (2h)
Jasper 4h Gathering articles for state-of-the-art (2 h), Discussing the subject (2h)
Wietske 6h Working on the wiki (0.5 h), Gathering articles for state-of-the-art (3.5 h), Discussing the subject (2h)

Week 2

Name Total Break-down
Daan 10h Working on the wiki (3 h), gathering articles for state-of-the-art (3 h), Discussing the subject (4h)
Job 10h Working on the wiki (2 h), gathering articles for state-of-the-art (4 h), Discussing the subject (4h)
Sanne 10h Working on the approach and planning (3h), working on the state of the art (3h) Discussing the subject (4h)
Jasper 10h Working on the wiki (2 h), gathering articles for state-of-the-art (4h) Discussing the subject (4h)
Wietske 17.5 h Working on the wiki (1 h), discussing the subject (4h), gathering articles for state-of-the-art (3 h), organizing sources on wiki by subject (1h), writing about the effect of music (1h), writing about social groups and alcohol(1.5 h), reading articles (6h)

Week 3

Name Total Break-down
Daan 14h Working on the wiki (7h), gathering articles for state-of-the-art (3), Discussing the subject (4h)
Job 12h Working on the wiki (6h), gathering articles for state-of-the-art (2h), Discussing the subject (4h)
Sanne 8h Working on the approach and planning (2), working on the state of the art (2) Discussing the subject (4h)
Jasper 12h Working on the wiki (2h), Discussing the subject (4h), setup Github(0.5h), Started programming simulation (5.5h)
Wietske 18h Working on the wiki (2 h), discussing the subject (4h), gathering articles for state-of-the-art (1 h), writing about factors in a bar (2 h), incorporating articles in state-of-the-art (2h), reading articles (4h), writing about alcohol (3h)

Week 4

Name Total Break-down
Daan 26h Discussing the subject (4h), improve/create code skeleton (4h), Alter data structure for more complete result collection (4h), create simulation flow + implement more club variables(2h), Experimented with exporting variables to MATLAB (2h), Create visualisation for crowd and bar objects (3h), Create crowd position/movement system (3h), Create collision system for movement (3h), implement alcohol vending and consumption (1h)
Job 14h Discussing the subject (4h), gathering articles for state-of-the-art (4h), working on the wiki (2h), working on java for Matlab (4h)
Sanne 4h Discussing the subject (4h)
Jasper 6h Discussing the subject (4h), Making small changes in the code (1h), working on the wiki(placed github link under deliverables)(1h)
Wietske 16h Discussing the subject (4h), Working on the wiki (2h), writing about alcohol (2h), Reading and processing articles (4h), writing about AI (2h), organising state-of-the-art (1h), contacting bar owners (1h)

Week 5

Name Total Break-down
Daan 10h Discussing the subject (4h), working on code clean-up (4h), reformatting/rewriting wiki elements (2h)
Job 16h Discussing the subject (4h), working on java for Matlab (8h), working on Matlab (4h)
Sanne 10h Discussing the subject (4h), Working on survey for stakeholders (2h), Classifying priority of variables (1h), working on java for foundational work (4h)
Jasper 9h Discussing the subject (4h), working on java for relationships between variables (5h)
Wietske 17h Discussing the subject (4h), Working on survey for stakeholders (1h), Processing results from survey for stakeholders (2h), Classifying priority of variables (1h), Gathering articles about missing information (4h), Processing articles about missing information (4h), working on the wiki (1h)

Week 6

Name Total Break-down
Daan 9h Discussing the subject (3h), reformatting/rewriting wiki elements (2h), Creating presentation draft (4h)
Job 14h Discussing the subject (3h), working on java for Matlab (4h), working on Matlab (4h), working on flowchart (3h)
Sanne 9h Discussing the subject (3h), working on java for state machine(6h)
Jasper 9h Discussing the subject (3h), working on java for relationships between variables (6h)
Wietske 13.5h Discussing the subject (3h), structurize wiki (0.5h), searching, reading, and processing articles (6h), working on the wiki (3h), add requirements (1h)

Week 7

Name Total Break-down
Daan 11h Discussing the subject (3h), updating wiki (mostly variables) (4h), checking consistency between wiki and program (4h)
Job 11h Discussing the subject (3h), updating wiki (2h), working in java (4h), working in Matlab (2h)
Sanne 13h Discussing the subject (3h), refactoring the code(8h), implementing states (2h)
Jasper 11h Discussing the subject (3h), reading the code after refactor(2h), Adding energy and updating the gui(3h), working on states(3h)
Wietske 17h Discussing the subject (3h), using research to determine conditions for state transitions (6h), adding variables (2h), searching and reading academic sources to support assumptions (4h), processing found sources (2h)

Week 8

Name Total Break-down
Daan 22h Discussing the subject(8h), adding discussion section (4h), discussing results (3h), adding result section (3h), working on presentation (2h), rereading/finalizing wiki (2h)
Job 22h Discussing the subject(8h), working in java and generating figures Matlab(9h), working on results(1h), working on wiki(2h),working on presentation(2h)
Sanne 25h Discussing the subject(8h),working on finalizing the simulation (10h), working on the presentation (5h),.working on the wiki (2h)
Jasper 19h Discussing the subject(8h), added talking state (2h), Working on slides (3h), Working on presentation(2h), finalizing code(4h)
Wietske 21h Discussing the subject(8h), beautifying state-of-the-art (2h), making research section report-ready (2h), started on discussion (1h), working on wiki (3h), writing experiments (2h), writing conclusion (3h)

References

  1. Mintel, 2002. Nightclubs. UK, Leisure Intelligence Pursuits, December 2002.
  2. E.F. Codd (1968). Cellular Automata, ACM Monograph Series. Academic Press, New York
  3. 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 Serena H. Chen, Anthony J. Jakeman, John P. Norton. (2008) Artificial Intelligence techniques: An introduction to their use for modelling environmental systems. https://doi.org/10.1016/j.matcom.2008.01.028
  4. F. Fdez-Riverola, J.M. Corchado. Improved CBR system for biological fore-casting, EOAI, Workshop 23, Binding Environmental Sciences and Artificial Intelligence, Valencia, Spain (2004)
  5. A. Aamodt, E. Plaza. Case-based reasoning: foundational issues, methodological variations, and system approaches. AI Commun., 7 (1994), pp. 35-59
  6. X. Yao. Evolving artificial neural networks. Proc. IEEE, 87 (9) (1999), pp. 1423-1447
  7. D.M. Rodvold, D.G. McLeod, J.M. Brandt, P.B. Snow, G.P. Mur-phy. Introduction to artificial neural networks: taking the lid off the black box. Prostate, 46 (2001), pp. 39-44
  8. M. Adya, F. Collopy. How effective are neural networks at forecasting and prediction? A review and evaluation. J. Forecasting, 17 (1998), pp. 481-495
  9. D.E. Goldberg. Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley Publishing Co., Reading, MA (1989)
  10. B.P. Buckeles, F.E. Petry. Genetic Algorithms. IEEE Computer Society Press, Los Alamitos, CA (1992)
  11. R. Sutton, A. Barto, Reinforcement Learning: An Introduction, http://www.cs.ualberta.ca/%7Esutton/book/ebook/the-book.html, 1998.
  12. L. Kaelbling, M. Littman, A. Moore. Reinforcement learning: a survey. J. Artif. Intell. Res., 4 (1996), pp. 237-285
  13. P. Abbeel, M. Quigley, A.Y. Ng, Using Inaccurate Models in Reinforcement Learning, http://ai.stanford.edu/∼ang/papers/icml06-usinginaccuratemodelsinrl.pdf, 2006.
  14. F. Hayes-Roth. Rule-based systems. Commun. ACM, 28 (1985), pp. 921-932
  15. K.C. Ng, B. Abramson. Uncertainty management in expert systems. IEEE Intell. Syst. Appl., 5 (1990), pp. 29-47
  16. V.R. Lesser. Multiagent systems: an emerging subdiscipline of AI. ACM Comput. Surv., 27 (1995), pp. 340-342
  17. 17.0 17.1 L. Parrott, R. Lacroix, K.M. Wade. Design considerations for the implementation of multi-agent systems in the dairy industry. Comput. Electron. Agric., 38 (2003), pp. 79-98
  18. R.A. Flores-Mendez, Towards a standardization of multi-agent system frameworks, ACM Crossroads 5, http://www.acm.org/crossroads/xrds5-4/multiagent.html, 1999.
  19. 19.0 19.1 B. Denby, S. Le Hégarat-Mascleb. Swarm intelligence in optimisation problems. Nucl. In-strum. Meth. A, 502 (2003), pp. 364-368
  20. E. Bonabeau, M. Dorigo, G. Theraulaz. Swarm Intel-ligence: From Natural to Artificial Systems. Oxford University Press, New York (1999)
  21. A. Gray, R. Kilgour, Frequently Asked Questions: Hybrid Systems, http://www.cecs.missouri.edu/∼rsun/hybrid-FAQ.html, 1997.
  22. C.V. Negoita. Expert Systems and Fuzzy Systems. Benjamin/Cummings Publishing Co., California (1985)
  23. I. Keramitsoglou, C. Cartalis, C.T. Tiranoudis. Automatic identification of oil spills on satellite images. Environ. Modell. Softw., 21 (2006), pp. 640-652
  24. C. Schmid, Course on Dynamics of Multidisplicinary and Controlled Systems, http://www.atp.ruhr-uni-bochum.de/rt1/syscontrol/main.html, 2005.
  25. R. Fuller. Introduction to Neuro-Fuzzy Systems. Physica-Verlag Heidelberg, New York (2000)
  26. G. Dounias, Hy-brid Computational Intelligence in Medicine, http://www.cs.queensu.ca/home/cisc875/Dounias_paper.pdf, 2003.
  27. AN AGENT-BASED APPROACH FOR OPTIMIZATION OF BARARRANGEMENTS WITHIN NIGHTLIFE VENUES. (2018) Bas Chatel, Alireza Faridamin, Mike la Grouw, Sara Neven and Viktor van der Valk
  28. Loud Music Is Scientifically Proven to Make You Drink More https://www.digitalmusicnews.com/2017/11/14/loud-music-drinking/
  29. 29.0 29.1 29.2 Guéguen, N., Jacob, C., Le Guellec, H., Morineau, T., & Lourel, M. (2008). Sound level of environmental music and drinking behavior: A field experiment with beer drinkers. Alcoholism:Clinical andExperimentalResearch,32 (10), 1795-1798.
  30. 30.0 30.1 Alcoholism: Clinical & Experimental Research. "Loud Music Can Make You Drink More, In Less Time, In A Bar." ScienceDaily. ScienceDaily, 21 July 2008. <www.sciencedaily.com/releases/2008/07/080718180723.htm>.
  31. 31.0 31.1 31.2 31.3 31.4 31.5 31.6 R. A. Knibbe, I. Van De Goor & M. J. Drop (1993) Contextual Influences on Young People's Drinking Rates in Public Drinking Places: An Observational Study, Addiction Research, 1:3, 269-278, DOI: 10.3109/16066359309005540
  32. Why Loud Music in Bars Increases Alcohol Consumption https://www.spring.org.uk/2008/09/why-loud-music-in-bars-increases.php
  33. 33.0 33.1 Sound Level of Background Music and Alcohol Consumption: An Empirical Evaluation August 1, 2004 https://doi.org/10.2466/pms.99.1.34-38
  34. https://learningmusic.ableton.com/make-beats/tempo-and-genre.html
  35. McElrea, H., & Standing, L. (1992). Fast music causes fast drinking. Perceptual and Motor Skills, 75 (2), 362.
  36. Milliman, R. E. (1986). The influence of background music on the behaviour of restaurant patrons. Journal of Consumer Research, 13(2), 286-9
  37. Samuel Joseph Down (2009). The effect of tempo of background music on duration of stay and spending in a bar. https://jyx.jyu.fi/bitstream/handle/123456789/20304/URN_NBN_fi_jyu-200905271640.pdf?sequence=1
  38. Hargreaves, D. J., & North, A. C. (Eds.). (1997). The social psychology of music. New York: Oxford University Press.
  39. 39.0 39.1 39.2 Jacob, C. (2006). Styles of background music and consumption in a bar: An empirical evaluation. International Journal of Hospitality Management,25 (4), 716–720.
  40. 40.0 40.1 40.2 Rutger C. M. E. Engels, Gert Slettenhaar, Tom ter Bogt, Ron H. J. Scholte (2011). Effect of Alcohol References in Music on Alcohol Consumption in Public Drinking Places. https://doi.org/10.1111/j.1521-0391.2011.00182.x
  41. The Impact of the Bass Drum on Human Dance Movement. Edith Van Dyck, Dirk Moelants, Michiel Demey, Alexander Deweppe, Pieter Coussement, Marc Leman. Music Perception: An Interdisciplinary Journal, Vol. 30 No. 4, December 2012; (pp. 349-359) DOI: 10.1525/mp.2013.30.4.349
  42. 42.0 42.1 42.2 42.3 42.4 42.5 Krzysztof Kubacki, Heather Skinner, Scott Parfitt, Gloria Moss (2007). Comparing nightclub customers’ preferences in existing and emerging markets. https://doi.org/10.1016/j.ijhm.2006.12.002
  43. Babor, T.F., Mendelson, J.H., Greenberg, I. et al. Experimental analysis of the ‘happy hour’: Effects of purchase price on alcohol consumption. Psychopharmacology 58, 35–41 (1978). https://doi.org/10.1007/BF00426787
  44. A Field Study of Bar-Sponsored Drink Specials and Their Associations With Patron Intoxication. (2009), Dennis L. Thombs, Ryan O'Mara, Virginia J. Dodd, Wei Hou, Michele L. Merves, Robert M. Weiler, Steven B. Pokorny, Bruce A. Goldberger, Jennifer Reingle, Chudley (CHAD) E. Werch.
  45. Bellizzi, J.A. and Hite, R.E. (1992), Environmental color, consumer feelings, and purchase likelihood. Psychology & Marketing, 9: 347-363. doi:10.1002/mar.4220090502
  46. 46.0 46.1 Hulshof, Bart (2013) The influence of colour and scent on people’s mood and cognitive performance in meeting rooms.
  47. D. Biswas, C. Szocs, R. Chacko, B. Wansink. Shining light on atmospherics: How ambient light influences food choices. Journal of Marketing Research, 54 (1) (2017), pp. 111-123
  48. C. Cajochen. Alerting effects of light. Sleep Medicine Reviews, 11 (6) (2007), pp. 453-464
  49. Y. Miwa, K. Hanyu. The effects of interior design on communication and impressions of a counselor in a counseling room. Environment and Behavior, 38 (4) (2006), pp. 484-502
  50. A. Steidle, L. Werth. In the spotlight: Brightness increases self-awareness and reflective self-regulation. Journal of Environmental Psychology, 39 (2014), pp. 40-50
  51. C.B. Zhong, V.K. Bohns, F. Gino. Good lamps are the best police: Darkness increases dishonesty and self-interested behaviour. Psychological Science, 21 (3) (2010), pp. 311-314
  52. L. Adams, D. Zuckerman. The effect of lighting conditions on personal space requirements. The Journal of General Psychology, 118 (4) (1991), pp. 335-340
  53. S.L. Neuberg, D.T. Kenrick, M. Schaller. Human threat management systems: Self-protection and disease avoidance. Neuroscience & Biobehavioral Reviews, 35 (4) (2011), pp. 1042-1051
  54. A. Mühlberger, M.J. Wieser, P. Pauli. Darkness-enhanced startle responses in ecologically valid environments: A virtual tunnel driving experiment. Biological Psychology, 77 (1) (2008), pp. 47-52
  55. 55.0 55.1 Seo Yoon Kang, Nara Youn, Heakyung Cecilia Yoon, The self-regulatory power of environmental lighting: The effect of illuminance and correlated color temperature, Journal of Environmental Psychology, Volume 62, 2019, Pages 30-41, ISSN 0272-4944, https://doi.org/10.1016/j.jenvp.2019.02.006.
  56. Leslie Adams & David Zuckerman (1991) The Effect of Lighting Conditions on Personal Space Requirements, The Journal of General Psychology, 118:4, 335-340, DOI: 10.1080/00221309.1991.9917794
  57. 57.0 57.1 Kuijsters, A., Silva, J., de Ruyter, B. E. R., & Heynderickx, I. (2014). Atmosphere perception: combining lightingand fragrance. In Y. A. W. de Kort, M. P. J. Aarts, F. Beute, A. Haans, I. E. J. Heynderickx, L. M. Huiberts, I.Kalinauskaite, P. Khademagha, A. Kuijsters, D. Lakens, L. van Rijswijk, A. C. Schietecat, K. C. H. J. Smolders,M. G. M. Stokkermans, ... W. A. Ijsselsteijn (Eds.), Proceedings EXPERIENCING LIGHT 2014: InternationalConference on the Effects of Light on Wellbeing, 10-11 November 2014, Eindhoven, The Netherlands (pp. 86-89). Eindhoven
  58. K. Yildirim, A. Akalin-Baskaya, M.L. Hidayetoglu, Effects of indoor color on mood and cognitive performance, Building and Environment, Volume 42, Issue 9, 2007, Pages 3233-3240, ISSN 0360-1323, https://doi.org/10.1016/j.buildenv.2006.07.037.
  59. 59.0 59.1 P. Valdez, A. Mehrabian, Effects of color on emotion, Journal of Environmental Psychology, 123 (4) (1994), pp. 394-409
  60. E. Grandjean, Ergonomics of the home, Wiley, New York (1973)
  61. 61.0 61.1 61.2 61.3 61.4 61.5 61.6 61.7 Gloria A. Moss, Scott Parfitt, Heather Skinner (2009). Men and Woman: Do They Value the Same Things in Mainstream Nightclubs and Bars? https://journals.sagepub.com/doi/abs/10.1057/thr.2008.37
  62. Schifferstein, H.N.J., Talke, K.S.S. & Oudshoorn, D. Can Ambient Scent Enhance the Nightlife Experience?. Chem. Percept. 4, 55 (2011). https://doi.org/10.1007/s12078-011-9088-2
  63. Kuntsche E., Gmel G. 2013. Alcohol consumption in late adolescence and early adulthood - where is the problem? https://serval.unil.ch/notice/serval:BIB_93F9E1EBAF0B
  64. 64.0 64.1 64.2 Geller, E.S., Russ, N.W. and Altomari, M.G. (1986), NATURALISTIC OBSERVATIONS OF BEER DRINKING AMONG COLLEGE STUDENTS. Journal of Applied Behavior Analysis, 19: 391-396. doi:10.1901/jaba.1986.19-391
  65. 65.0 65.1 65.2 65.3 65.4 Heather Skinner, Gloria Moss and Scott Parfitt (2005). Nightclubs and bars: what do customers really want? The Business School, University of Glamorgan, Pontypridd, UK
  66. On the Impact of Mood on Behavior: An Integrative Theory and a Review (2000), Guido H.E. Gendolla, https://doi.org/10.1037/1089-2680.4.4.378
  67. Meryl Paula Gardner, Mood States and Consumer Behavior: A Critical Review, Journal of Consumer Research, Volume 12, Issue 3, December 1985, Pages 281–300, https://doi.org/10.1086/208516
  68. 68.0 68.1 Mood and representations of behaviour: The how and why. (2005) Camiel J. Beukeboom and Gün R. Semin. DOI:10.1080/02699930500203369
  69. The Role of Mood States in Self-Regulation: Effects on Action Preferences and Resource Mobilization. Guido H.E. Gendolla, Kerstin Brinkmann. Published online: September 1, 2006. https://doi.org/10.1027/1016-9040.10.3.187
  70. 70.0 70.1 Mood States and Prosocial Behavior (1989), Peter Salovey and David L. Rosenhan
  71. 71.0 71.1 Weyant, J. M. (1978). Effects of mood states, costs, and benefits on helping. Journal of Personality and Social Psychology, 36(10), 1169–1176. https://doi.org/10.1037/0022-3514.36.10.1169
  72. 72.0 72.1 72.2 72.3 72.4 Geoffrey Hunt, Molly Moloney & Adam Fazio (2014) “A Cool Little Buzz”: Alcohol Intoxication in the Dance Club Scene, Substance Use & Misuse, 49:8, 968-981, DOI: 10.3109/10826084.2013.852582
  73. Monk R. L., Heim D. (2014). A systematic review of the alcohol norms literature: a focus on context. Drugs Educ Prev Policy 2014; 21: 263–282. https://www.tandfonline.com/doi/full/10.3109/09687637.2014.899990
  74. Senchak, M., Leonard, K. E., & Greene, B. W. (1998). Alcohol use among college students as a function of their typical social drinking context. Psychology of Addictive Behaviors, 12(1), 62–70. https://doi.org/10.1037/0893-164X.12.1.62
  75. Dunbar, R.I.M., Launay, J., Wlodarski, R. et al. Functional Benefits of (Modest) Alcohol Consumption. Adaptive Human Behavior and Physiology 3, 118–133 (2017). https://doi.org/10.1007/s40750-016-0058-4
  76. Dunbar, R., Duncan, N., & Nettle, D. (1995). Size and structure of freely forming conversational groups. Human Nature, 6, 67–78.
  77. Dezecache, G., & Dunbar, R. (2012). Sharing the joke: the size of natural laughter groups. Evolution & Human Behavior, 33, 775–779.
  78. Dunbar, R. I. M. (2016). Sexual segregation in human conversations. Behaviour, 153, 1–14.
  79. Krems, J. A., Dunbar, R. I. M., & Neuberg, S. L. (2016). Something to talk about: are conversation sizes constrained by mental modeling abilities? Evolution and Human Behavior, 37, 423–428.
  80. Dahmardeh, M. & Dunbar, R. I. M. (2017). What shall we talk about in Farsi? Content of everyday conversations in Iran. Evolution and Human Behavior, in press.
  81. Kairouz S., Gliksman L., Demers A., Adlaf E. M (2002). For all these reasons, I do... drink: a multilevel analysis of contextual reasons for drinking among Canadian undergraduates. J Stud Alcohol 2002; 63: 600–608. https://www.jsad.com/doi/abs/10.15288/jsa.2002.63.600
  82. 82.0 82.1 82.2 Thrul J., Kuntsche E. (2015). The impact of friends on young adults’ drinking over the course of the evening—an event‐level analysis. Addiction 2015; 110: 619–626.
  83. 83.0 83.1 83.2 P. P. AITKEN, AN OBSERVATIONAL STUDY OF YOUNG ADULTS' DRINKING GROUPS—II. DRINK PURCHASING PROCEDURES, GROUP PRESSURES AND ALCOHOL CONSUMPTION BY COMPANIONS AS PREDICTORS OF ALCOHOL CONSUMPTION, Alcohol and Alcoholism, Volume 20, Issue 4, 1985, Pages 445–457, https://doi.org/10.1093/oxfordjournals.alcalc.a044569
  84. 84.0 84.1 Johannes Thrul, Florian Labhart, Emmanuel Kuntsche (2016). Drinking with mixed‐gender groups is associated with heavy weekend drinking among young adults (2016): https://onlinelibrary.wiley.com/doi/full/10.1111/add.13633
  85. Akanidomo K. J. Ibanga, Victor A. O. Adetula, Zubairu K. Dagona (2009). Social Pressures to Drink or Drink a Little More: The Nigerian Experience. https://journals.sagepub.com/doi/abs/10.1177/009145090903600107
  86. Pia Mäkelä, Antti Maunu (2016). Come on, have a drink: The prevalence and cultural logic of social pressure to drink more. https://doi.org/10.1080/09687637.2016.1179718
  87. The modeling of alcohol consumption: a meta-analytic review. B M Quigley and R L Collins Journal of Studies on Alcohol 1999 60:1, 90-98
  88. 88.0 88.1 Jakob Demant & Jukka Törrönen (2011) Changing Drinking Styles in Denmark and Finland. Fragmentation of Male and Female Drinking Among Young Adults, Substance Use & Misuse, 46:10, 1244-1255, DOI: 10.3109/10826084.2011.569965
  89. 89.0 89.1 89.2 Sander M. Bot, Rutger C.M.E. Engels, Ronald A. Knibbe, Wim H.J. Meeus (2007). Pastime in a pub: Observations of young adults' activities and alcohol consumption. https://doi.org/10.1016/j.addbeh.2006.05.015
  90. 90.0 90.1 Aniko Maraz, Orsolya Király, Róbert Urbán, Mark D. Griffiths, Zsolt Demetrovics (2015). Why Do You Dance? Development of the Dance Motivation Inventory (DMI) https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0122866
  91. 91.0 91.1 Jade Boyd (2014) ‘I go to dance, right?’: representation/sensation on the gendered dance floor, Leisure Studies, 33:5, 491-507, DOI: 10.1080/02614367.2013.798348
  92. J. Leiper, Thirst Physiology, Editor(s): Benjamin Caballero, Encyclopedia of Human Nutrition (Third Edition), Academic Press, 2013, Pages 280-287, ISBN 9780123848857, https://doi.org/10.1016/B978-0-12-375083-9.00267-1.
  93. 93.0 93.1 93.2 93.3 93.4 93.5 93.6 The Alcohol Pharmacology Education Partnership https://sites.duke.edu/apep/module-2-the-abcs-of-intoxication/content-the-blood-alcohol-concentration-bac-estimates-the-degree-of-intoxication/ retrieved on 01-03-2020
  94. 94.0 94.1 94.2 94.3 William C. Kerr, Thomas K. Greenfield, Lorraine T. Midanik (2006) How many drinks does it take you to feel drunk? Trends and predictors for subjective drunkenness. https://doi.org/10.1111/j.1360-0443.2006.01533.x
  95. Dr. Lorraine T. Midanik (2003) Definitions of Drunkenness, Substance Use & Misuse, 38:9, 1285-1303, DOI: 10.1081/JA-120018485
  96. Mitchell, M.C., Jr., Teigen, E.L. and Ramchandani, V.A. (2014), Absorption and Peak Blood Alcohol Concentration After Drinking Beer, Wine, or Spirits. Alcohol Clin Exp Res, 38: 1200-1204. doi:10.1111/acer.12355
  97. 97.0 97.1 97.2 97.3 97.4 George E. Vaillant and Mark Keller (2020). Alcohol consumption (Encyclopædia Britannica). https://www.britannica.com/topic/alcohol-consumption ,accessed on March 03, 2020
  98. Paton Alex. Alcohol in the body BMJ 2005; 330 :85
  99. Paton Alex. Alcohol in the body BMJ 2005; 330 :85
  100. John D. Clapp, Mark B. Reed, Jong W. Mina, Audrey M. Shillington, Julie M. Croff, Megan R. Holmes, Ryan S. Trim (2009). Blood alcohol concentrations among bar patrons: A multi-level study of drinking behavior https://doi.org/10.1016/j.drugalcdep.2008.12.015
  101. Paton Alex. Alcohol in the body BMJ 2005; 330 :85
  102. Vogel-Sprott M (1967) Alcohol effects on human behavior under reward and punishment. Psychopharmacologia 11:337–344
  103. Glowa JR, Barrett JE (1976) Effects of alcohol on punished and unpunished responding of squirrel monkeys. Pharmacol Biochem Behav 4:169–173
  104. Vogel RA, Frye GD, Wilson JH, Kuhn CM, Kuepke KM, Mailman RB, Mueller RA, Breese GR (1980) Attentuation of the effects of punishment by ethanol: comparisons with chlordiazepoxide. Psychopharmacology 71:123–129
  105. Josephs RA, Steele CM (1990) The two faces of alcohol myopia: attentional mediation of psychological stress. J Abnorm Psychol 99:115–126
  106. Lane, S.D., Cherek, D.R., Pietras, C.J. et al. Alcohol effects on human risk taking. Psychopharmacology 172, 68–77 (2004). https://doi.org/10.1007/s00213-003-1628-2
  107. DAVID L. McMILLEN and ELISABETH WELLS-PARKER (1987). THE EFFECT OF ALCOHOL CONSUMPTION ON RISK-TAKING WHILE DRIVING https://doi.org/10.1016/0306-4603(87)90034-7
  108. 108.0 108.1 Richards, Jerry B and Zhang, Lan and Mitchell, Suzanne H and Wit, Harriet (1999) DELAY OR PROBABILITY DISCOUNTING IN A MODEL OF IMPULSIVE BEHAVIOR: EFFECT OF ALCOHOL, Journal of the Experimental Analysis of Behavior, 2:71, 121—143, DOI: 10.1901/jeab.1999.71-121
  109. 109.0 109.1 Catherine N. M. Ortner, Tara K. MacDonald, Mary C. Olmstead, ALCOHOL INTOXICATION REDUCES IMPULSIVITY IN THE DELAY-DISCOUNTING PARADIGM, Alcohol and Alcoholism, Volume 38, Issue 2, March 2003, Pages 151–156, https://doi.org/10.1093/alcalc/agg041
  110. 110.0 110.1 110.2 Pete Seaman and Theresa Ikegwuonu (2010). Young people and alcohol: influences on how they drink http://www.ias.org.uk/uploads/pdf/Young%20people/alcohol-young-adults-summary.pdf
  111. 111.0 111.1 Rutger C. M. E. Engels, Ronald A. Knibbe & Maria J. Drop (2009). Visiting Public Drinking Places: An Explorative Study into the Functions of Pub-Going for Late Adolescents https://doi.org/10.3109/10826089909039408
Personal tools