Collection Mémoires et thèses électroniques
Accueil À propos Nous joindre

CHAPTER 4: Identification of the Application Domain, the Users’ Needs, the Characteristics of the Phenomenon to be Simulated and its Environment

Table des matières

This chapter presents the first two steps of the generic method that we propose in this thesis. The first step aims to identify the geosimulation users’ needs, while the second aims to specify the characteristics of the phenomenon to be simulated. This chapter also presents the illustration of these two steps using the customers’ shopping behavior in a mall as a case study.

Geosimulation is a kind of computer-based simulation which is concerned with the design and construction of high-resolution spatial models of phenomena in order to understand how they behave and operate in geographic environments (Benenson and Torrens, 2004). The simulated phenomena can be natural or artificial, human or not. Geosimulation is generally used to explore ideas and hypotheses about how phenomena operate and to solve problems in geographic contexts. In the geosimulation field, applications are generally used for decision-making purposes. These decisions are extremely influenced by the spatial characteristics of the phenomenon to be simulated, or by the geographic features of the environment in which the phenomenon occurs. In order to develop useful geosimulation applications, it is relevant to identify the end-users of these applications and to specify their needs. Hence, the importance of our method’s first step: ‘ identification of the application domain, the geosimulation end-users, and their needs ’. In this first step, we attempt to answer the following questions: what do we want to simulate ?, where do we want simulate ?, who are the users of the simulation ?, and why the users want to use the simulation ?.

In this first step, we define the geosimulation application domain: The phenomenon to be simulated and its environment. If we want to simulate this phenomenon on a computer, we must learn about it in order to know how it operates and behaves in its geographic environment. Learning about the phenomenon to be simulated is the aim of the second step of our proposed method which is called: ‘ identification of the characteristics of the phenomenon to be simulated and its environment ’.

This chapter presents in detail these two steps and illustrates them using the customers’ shopping behavior in a mall as a case study. It is organized as follows: Section 4.1 presents generic descriptions of the two first steps of our proposed method. Section 4.2 aims at illustrating the method’s first step by presenting the intended end-users of the shopping behavior multiagent simulator. It also summarizes the main needs of these users. Then, in Section 4.2, we present the illustration of the second step of the method. Hence, we show the results of the shopping behavior study. This in-depth study involved several disciplines related to the shopping behavior in a mall, and to other similar behaviors. Among these disciplines we can cite: discipline dealing with shopping/consumption/buying behaviors, marketing, social behavior, social psychology, etc. Next, Section 4.3 aims at specifying the characteristics and the basic concepts of human shopping behavior in a mall. Section 4.4 then presents shopping in a mall as a social activity. It presents a literature review and specific results related to the shopping activity in groups, and to the structure and behavior of groups in general. Section 4.5 concludes by discussing the issues presented in the chapter.

This section presents generic descriptions of the first two steps of our proposed method and defines some basic concepts.

The first step of our proposed method (see Fig. 4.1) aims at:

(1) identify the application domain : We identify the phenomenon to be simulated and the simulated environment. In this sub-step, we attempt to answer the following question: What do we want simulate and where ?;

(2) identify the geosimulation’s end-users : We identify the future users of the geosimulation application. In this sub-step, we attempt to answer the following question: Who are the end-users of the geosimulation application ?;

(3) identify the main needs of these users : Once the geosimulation end-users are defined, it is relevant to identify their needs concerning the use of the geosimulation application. In this sub-step, we attempt to answer the following question: Why the end-users need the geosimulation application ?;

(4) identify the geographic limits of the simulated environment : Since we deal with geosimulation, the environment, and especially, its geographic features, influence the behavior of the phenomenon to be simulated. Hence, in order to reduce the complexity of the geosimulation, it is relevant to identify the geographic limits of the simulated environment. In this sub-step, we attempt to answer the following question: What are the geographic limits of the simulation environment ?

As outputs of this step, we have: description of the application domain of the geosimulation, geosimulation future end-users, description of the end-users’ needs, and describing the geographic limits of the simulated environment.

In the previous step, we identified some elements related to the geosimulation application domain (the simulated phenomenon and its environment), the intended users, and their needs. If we want to develop operational computer-based geosimulation of the phenomenon, we must specify its main characteristics and those of its environment, taking into account the elements defined in the previous step. The second step of our method aims at defining these characteristics and contains the following sub-steps (Fig. 4.2):

(1) Learn about the phenomenon to be simulated and its environment : Before identifying the characteristics of the phenomenon to be simulated, we must learn about it. Therefore, we can (i) study documents concerning it, (ii) observe it when it behaves (if it is possible), (iii) consult experts or specialists about it, etc. It is important to notice that learning about a phenomenon and its environment can involve several disciplines and science areas.

(2) Identify the characteristics of the phenomenon to be simulated : Based upon the study resulting from the previous sub-step, we can identify the characteristics of the phenomenon to be simulated. To characterize a phenomenon, we must identify its structure and behavior. To identify its structure, it is important to identify the internal elements, factors or variables that characterize that structure (can come from the system) and that influence its behavior. To identify its behavior, we must characterize the processes that compose this behavior. Since we deal with geosimulation, it is important to take into account the spatial aspects of the phenomenon when characterizing its characteristics.

(3) Identify the characteristics of the simulation environment : When we deal with geosimulation, the environment, and especially, its geographic features, are important. In this sub-step, we identify the characteristics of the simulation environment within which the phenomenon to be simulated behaves. Hence, we must identify the structures and behaviors of each element belonging to the simulation environment taking into account the geographic limits of such environment.

(4) Design an initial model of the phenomenon to be simulated and its environment : This sub-step aims to design an initial version of the model of the phenomenon and its environment. The design of such model is based upon the characteristics of the phenomenon to be simulated and its environment. In this model, we represent the structure of the phenomenon to be simulated, the processes composing its behavior, the structure and behaviors of the elements composing the simulation environment, and the interactions of the phenomenon to be simulated and its environment. This model aims to describe, graphically, the phenomenon and the environment to be simulated in order to better understand them.

In sections 4.3 to 4.5, we aim at illustrating, respectively, these steps using customer’s shopping behavior in a mall as a case study.

Like any other software or program, multiagent geosimulations are intended to be used by end-users in order to solve problems and make informed decisions. In multiagent geosimulation applications, decisions are influenced by the spatial features of the phenomenon to be simulated, or by the geographic characteristics of its environment.

In the case of the shopping behavior domain, we can identify:

- The application domain - the shopping behavior in a mall (shoppers).

We distinguish two kinds of shoppers: individual shoppers which come to the mall alone, groups of shoppers (which can be couples, groups of friends or colleagues, families, etc.) and a crowd of shoppers that is composed of all the individual shoppers and groups of shoppers within the mall. Hence, the phenomenon to be simulated is composed of individual shoppers and groups of shoppers, while the simulated environment is ‘the mall’.

- The users of the geosimulation application - the mall managers:

Mall managers are the primary users of our prototype system for shopping behavior multiagent geosimulation. They will use it to better understand the spatial behaviors of the shoppers visiting the mall and to assess the impact of the spatial layout of the mall on these behaviors.

- The main needs of the users :

The main need of mall’s managers is to have access to an efficient simulation prototype that can simulate customers’ shopping behaviors in a mall. Mall’s managers are particularly interested in understanding which particular categories of shoppers visit specific places (particular stores, kiosks, meeting areas, etc.) and which paths they follow in the mall. They are also interested in better understanding the characteristics of shoppers’ flows in the mall. As examples of indicators of interest to mall managers, we can mention: the number of shoppers entering by a specific door or going through a specific corridor in the mall, the genders and age categories of shoppers going through a corridor, the category of stores visited by particular categories of the shoppers , etc. When changing the configuration of the virtual mall in the simulation prototype, mall mangers want to observe and assess how the changes may influence the behavior of the virtual shoppers, the evaluation being quantified thanks to the chosen indicators. As examples of modifications that can be carried out in the virtual mall, mall managers can exchange the positions of two places in the mall (stores, kiosks, washrooms, etc.), they can close a corridor or a door, change the position of a door in the mall. It is interesting for mall managers to have an idea of the impacts of these changes, using the geosimulation in the virtual mall, before carrying out these changes in the real mall, changes which may be very costly.

Since the phenomenon to be studied has been defined (individual shoppers and groups of shoppers), the simulated environment is defined (the mall), end-users of the shopping behavior simulator are identified, and their needs are characterized, we can start to study and learn about the shopping behavior in a mall. This study is conducted in several scientific disciplines related to the shopping behavior in a mall, or to other behaviors similar to the aforementioned. The study results are presented in Sections 4.4 (individual shoppers) and 4.5 (groups of shoppers) of the chapter.

The first step of our method has been presented in the previous section. This section illustrates the second step whereby the characteristics of the shopping behavior in a mall are identified. Before doing this though, we need to learn more about it. Sub-section 4.4.1 presents the results of an in-depth literature review concerning the shopping behavior in a mall. Based upon this literature review, we then identify the characteristics of the shopping behavior in Sub-section 4.4.2.

The shopping behavior is defined by (Haynes et al., 1994) as follows: ‘ how individuals or groups choose a store for shopping in a mall ’. Store choice and shopping patterns are based upon customers’ perception, images, and attitudes formed from experiences, information, and needs. Furthermore, the shopping behavior involves a decision process related to where consumers shop, how they shop, and what they purchase. This decision process is often initiated by shopping behavior motives, which determine why consumers shop and make purchases at certain retail stores (Moschis, 1992)(Stafford and stafford, 1986). As stated by (Haynes et al., 1994), the shopping behavior involves three basic components: The stores’ attributes ( where ), the consumers’ characteristics ( who ) and the choice context ( how and why ).

The number of studies concerning shopping malls and shopping behavior is very limited (Wakefield and Baker, 1998). Moreover, shopping behavior research frequently concentrates its efforts on individual stores, and not on the mall itself. Furthermore, most research in this area dates back to the 1970s, 1980s, and the early half of the 1990s (Reynolds et al., 2002). What’s more, most such studies focused on the shopper ( who ) and its demographic variables (Stone, 1954) (Smith, 1956). These studies gave a detailed taxonomy of customers, but they could not explain why people shop (Hassay and Smith, 1996). Some researchers focused on shoppers in a mall according to the ‘ how ’ question of shopping behavior instead of ‘ who ’ they are. For example, (Roberts and Merrilees, 2001) presented a categorization of the shopping activities based upon the mall components. Hence, we can find: retail-based activities (buy items and brands), service-based activities (buy services), entertainment activities (go to the mall for entertainment and play), and social activities (go to the mall to socialize and meet people). (Bloch et al., 1994) developed a segmentation model that clusters customers by the shopping activity they perform in a mall, or not (going to a shop, going to a bank, having a snack, etc.). The taxonomy of shopping activities presented by (Bloch et al., 1994) is composed of 13 activities. Their method represents a significant improvement on previous models because it is based upon behavioral variables (action), and not merely descriptive variables such as age and gender. (Bloch et al., 1994) talked about the ‘ where ’ by acknowledging the importance of the environment and atmospheric factors that affect the shopping behavior: ‘ additional research on the environmental psychology of malls using different measures and methods seems highly worthwhile given the substantial resources being devoted to mall design and rehabilitation ’ (Bloch et al., 1994).

In order to better simulate the shopping behavior in a mall we need to understand all the aspects of this behavior. This means we need to better understand:

- Who is doing the shopping: shopper characteristics;

- Why she/he is doing the shopping: motivations for the shopping;

- How she/he is doing the shopping: how the shopper accomplishes the shopping behavior: what are the shopping activities that can be done in the mall, how the shopper makes shopping decisions, etc.;

- Where the shopper does his/her shopping: shopping environment characteristics (the mall) and the shopping atmosphere.

The who , why, and how questions come from the shoppers, while the where question comes from the shopping mall.

In addition, we notice a noteworthy lack in terms of empirical research works dealing with the shopping behavior in a mall (Wakefield and Baker, 1998). In our study, we need such empirical research works in order to simulate the shopping behavior in a mall. Fortunately, there are a number of works in the literature that deal with some human behaviors that are very similar to the shopping behavior in a mall. These behaviors are the consumption behavior and the shopping behaviors in stores .

(Fotheringham, 1998) argued that in shopping behavior in a mall, the choice of the store is classified primarily as a cognitive process. ‘Store choice behavior has been found to be similar to items/brand choice by a consumer behavior. The only difference being the importance of the geographic dimension. While the brand choice is devoid of any geography, the choice of a store is very much influenced by its location ’. This dissertation shows the similarity between the shopping behavior in a mall (the shopper decides which store to visit in order to enter and purchase some items or services), and the shopping behavior in a store (the shopper decides to purchase an item or brand inside a store).

In the consumer behavior discipline, (Perner, 2005) defines the consumption behavior as ‘ the study of individuals, groups, or organisations, and the process they use to select, secure, use, and dispose of products, services, experiences, or ideas to satisfy needs, and the impact that these processes have on the consumer and society ’. Another definition given by (James et al., 1973) is « acts of individuals directly involved in obtaining and using economic goods and services, including the decision processes that precede and determine these acts ». These definitions show the similarity between the consumption behavior and the shopping behavior in a mall.

Given the high similarity between the shopping behavior in a mall, and both the consumption behavior (in the consumer behavior discipline) and the shopping behavior inside a store , we can characterize the shopping behavior in a mall based upon the study results and research works of these other fields of research.

Shopping behavior in a mall is similar to consumption behavior. In order to understand the human shopping behavior in a mall, it is relevant to study the human consumption behavior. This sub-section presents our literature review of consumer behavior, with the goal to understand some aspects of the shopping behavior in a mall.

It is well accepted in the literature concerning consumer behavior that the consumption behavior is influenced by several factors, and is composed by various processes. In the following sub-sections, we present in detail some of these factors and processes.

(Duhaime et al., 1996) argued that the best way to study human consumption behavior is to find « Why they act like they do ?». The majority of the researchers who study this subject are interested to know why humans buy some goods and adopt such consumption behavior, etc. These researchers emphasize that consumption behavior is influenced by two types of factors: internal factors (e.g., needs, values, lifestyle, personality, attitudes, gender, lifecycle, etc.), and environmental and social factors (e.g., culture, social class, reference groups, family, etc.) (Duhaime et al., 1996). Other factors, which are cited in the literature such as emotion, humour, time, advertisement, etc., are presented as additional factors or miscellaneous . They are discussed in the following points.

Internal factors :

Internal factors contain the majority of factors that come from the consumer, and that influence his consumption behavior. The main factors are:

Motivation and needs : Behavior is initiated through motivations and needs (Bayton, 1958). The author argues that motivation is one of the basic factors in consumer behavior. Motivation arises out of tension-systems which create a state of disequilibrium in the mind of the individual (Helgeson et al. , 1984). This triggers a sequence of psychological events directed toward the selection of a goal, which the individual anticipates will bring about relief from such tension, and the selection of patterns of action which he anticipates, will bring about the achievement of the goal. To better understand consumption behavior, we need to know the fundamental needs and motivations, which are the origin of this behavior (Duhaime et al., 1996). Some researchers tried to identify the needs systems of humans. For example, (Maslow, 1954) proposed a theory, with a five-level-hierarchy of human needs (see Fig 4.3). This hierarchy is now central to much thinking in consumption behavior.

(Maslow, 1954) was not the only theorist to focus his efforts on human needs as the motivating force behind human behavior. For example, (Murray, 1938) elaborated a typology of physiological needs even more detailed than that proposed by Maslow.

Values, beliefs, and attitudes : Values, beliefs, and attitudes greatly affect human behavior (Duhaime et al., 1996). A value is a standard that guides one’s actions, attitudes, comparisons, evaluations, and justifications toward oneself and others (Rokeach, 1960). While motivations and needs of human behavior continually change, values and beliefs are relatively permanent (Rokeach, 1960). Rokeach developed what he called ‘ Rokeach Value Survey: RVS ’, which is an instrument to operationalize the value concept. Researchers closely compare the concepts of attitudes and beliefs when studying values. The idea of studying attitudes and beliefs, when learning about values, is important because the three concepts are closely related to one another. An attitude is a relatively enduring organization of beliefs around an object or situation, predisposing one to respond in some preferential manner (Rokeach, 1960). An attitude is thus, a package of beliefs consisting of interconnected assertions to the effect that certain things about a specific situation are true or false, and other things about it are desirable or undesirable (Morris, 1956). A belief is any simple portion of knowledge, conscious or unconscious, inferred from what a person says or does (Weiner, 1998). According to a study by (Weiner, 1998), all beliefs are a predisposition to action, and an attitude is thus, a set of interrelated predispositions to action, organized around an object.

Lifestyle : (Cosmas, 1982) argued that the total assortment of goods and services used by a consumer is hypothesized to be a mirror image of his/her lifestyle. Lifestyle refers to the distinctive ways in which consumers live, how they spend their time and money, and what they consider important- activities, interests, and opinions.

Perception : Perception deals with recognizing, selecting, organizing, and interpreting stimuli in order to make sense of the world around us. People receive stimuli from their environment through the five senses, which they then must interpret. People are selective and interpret stimuli that reinforce and enhance their existing beliefs. Consumers tend to interpret what they perceive in such a way so that it does not conflict with their basic attitudes, personalities, motives, or aspirations. They pay attention to stimuli deemed relevant to existing needs, wants, beliefs, and attitudes, and disregard the rest.

Knowledge and learning : Consumer knowledge and learning significantly affect consumption behavior (Craik and Watkins, 1973). Information processing describes the series of steps, by which information (or stimulus) is encountered through some exposure to a person’s senses, interpreted, understood and accepted, and stored in memory for future use in making of decisions (McGuire, 1976). Elaboration, via mental processing, transforms this information into beliefs, attitudes, and intentions that determine product choice and related aspects of purchase. The degree of integration between the stimulus and existing knowledge that occurs while a stimulus is being processed, will influence the amount of learning that takes place (Craik and Watkins, 1973).

Self-concept : (Sirgy, 1982) presented the influence of the self-concept on consumption behavior. Self-concept is an organized set of perceptions of the self, comprised of such elements as the perceptions of one’s characteristics and abilities; the perception of oneself in relation to others; and objectives, goals, and ideals that are perceived as either positive or negative (Rogers, 1951).

Personality : One of the more engrossing concepts in the study of consumer behavior is that of personality. Personality accounts for consistent patterns of behavior based upon enduring psychological characteristics (Kassarjian, 1971). It is the pattern of traits and behaviors that makes one individual unique and different from all others.

Environmental and social factors :

Consumers do not live or make decisions in isolation. The values, beliefs, and opinions of those who surround the consumer affect his/her decisions. Among the environmental and social influences on the consumer are: culture, sub-culture, social class, reference groups, and family.

Demographic variables : Demographic variables have a long history in marketing and consumer behavior disciplines (Kalyanam and Putler, 1994).(Kalyanam and Putler, 1994) presented the importance of taking into account the demographic variables in human consumption behavior. The most important variables studied by the researchers of consumer behavior are the following: gender; age group; marital status; occupation and employment sector; habits, and preferences.

Culture and sub-culture : Culture consists of a society’s beliefs, values, ethics, customs, shared meanings, rules, rituals, norms, and traditions. Culture provides people with a sense of identity and an understanding of acceptable behavior. Culture is deep-seated and enduring, but does change slowly over time (Clark, 1990). Sub-cultures are racial, ethnic, religious, or other groups whose members are distinguishable from the general population and who are held together by common culture and/or genetic ties. To the degree that people in an ethnic group share common customs, values, rituals, and traditions that are different from those of other ethnic groups or the larger society, they constitute a distinct ethnic group (Hirschman, 1982).

Social classes and stratification : Social stratification represents the hierarchical division of members of a society into relative levels of prestige, status, and power (Rossides, 1990). Social class refers to divisions, based upon economic and demographic characteristics. Those in the same stratum have roughly similar consumption, lifestyle, and income, and socialize with each other (Gilbert and Kahl, 1982).

Reference groups : Reference groups are individuals or collections of people whom the individual uses as a source, or point of comparison for attitudes, beliefs, values, or behaviors. Consumers belong to some of the groups that influence their consumer behavior, and either aspire to join, or work to avoid association with others. Some of these groups are formal groups and others are simply informal groups of friends (Homans, 1961).

Family : The most influential reference group is the consumer’s family (Qualls, 1982). The family teaches the consumer cultural values that have a substantial impact upon consumption behavior. It continues to be a point of reference, even when the consumer has formed his own household.

Additional factors (miscellaneous) :

Consumer behavior is influenced by other additional factors. These factors are presented in the following points:

Emotion : A large number of research works presented emotions as an important component of consumer behavior (Richins, 1997)(Lemoine, 2001). In their studies of consumption-related emotions, consumer behavior scholars have based much of their work on frameworks of emotion developed in psychology. The foundation laid by theorists in this field has provided a useful starting point for the investigation of emotions in the consumer behavior field (Richins, 1997). Some scholars have attempted to order the universe of emotions by identifying a set of basic fundamental emotions, although there is no widespread agreement concerning the number or the nature of such basic emotions. The most important scale in the domain of consumer behavior by marketing scholars is the PAD (Pleasure-arousal-dominance) emotion scale developed by (Mehrabian and Russell, 1974). Another interesting emotion scale often used in the sphere of consumer behavior, is the scale proposed by (Richins, 1997). This scale, called Consumption Emotion Scale (CES) (see Table 4.1), is specific to consumption behavior.

The mood : Mood has been described as a phenomenological property of a person’s subjectively perceived affective state. It is viewed as a mild, transient, generalized, and pervasive affective state, not an intense emotion, and not directed at specific target objects (Swinyard, 1993). A positive mood seems to make one kinder, more generous, more resistant to temptation, and more willing to delay self-rewards. Negative moods have produced greater dislike for peers, less volunteering for helping tasks, smaller contributions to charity, less resistance to temptation, and less willingness to delay self-rewards. However, the observed effects of negative moods have been less consistent than those of positive moods. Mood has been shown to have significant effects on consumer behavior (Swinyard, 1993). (Gardner, 1985) observed that the effects of mood may have a special impact on retail or service encounters because of their interpersonal nature, a view also supported by others (Swinyard, 1993). Studies of mood in shopping situations showed its effect on shopping behavior or intentions.

The role of involvement : The involvement has a greater effect when the shopping experiences are personally relevant to consumers or are «self-related or in some way instrumental in achieving their personal goals» (Swinyard, 1993). A large number of studies have reported that consumers, involved in a situation or product, are more active processors of related cognitive information. Involved consumers should attend to, and comprehend more information about a shopping situation, and should produce more elaborate meanings and inferences about it.

The consumer buying process is a complex matter, since many internal and external factors have an impact on the consumer’s buying decisions. When purchasing a product, researchers identified several processes through which consumers go (Duhaime et al., 1996) (see Fig 4.4). These processes are discussed in the following points.

Needs or problem recognition process :

Consumer decision making comes about as an attempt to solve consumer problems. A problem refers to ‘ a discrepancy between a desired state and an ideal state, which is sufficient to arouse and activate a decision making process ’ (Duhaime et al. , 1996). Consumers often note problems by comparing their current, or actual, situation, explicitly or implicitly, to some desired situation. Problems come in several different types. A problem may be an active one (e.g., you have a headache and would like as quick a solution as possible) or inactive -- you are not aware that your situation is a problem (e.g., a consumer is not aware that he or she could have more energy taking a new vitamin). Problems may be acknowledged (e.g., a consumer is aware that his or her car does not accelerate well enough), or unacknowledged (e.g., a consumer will not acknowledge that he or she consumes too much alcohol). Finally, needs can be relatively generic , as in the need for enjoyment (which can be satisfied in many different ways), or specific , as in the need to eat a chocolate ice-cream (Bettman and Park, 1980).

Information search process :

When the consumer recognizes a need (or problem), he/she starts to search for information about the items or services that can be used in order to satisfy that need. There are two principal approaches to searching: internal and external. Internal searches ( via the memorization search process ) are based upon what consumers already know (what is in his/her memory). A problem is that some items or services that can satisfy the need or solve the problem, are not remembered, or have never been heard of, and are therefore, not considered. In this case, the consumer can use external searches ( via the perception process ) that get people to either speak to others (getting information by word of mouth) or use other sources (such as advertisements or yellow-page listings) (Bettman, 1979; Punj, 1987). In the search processes, consumers often do not consider all possible alternatives. Some are not known (the « unawareness » set ), some were once known, but are not readily accessible in the memory (the « inert » set ), while others are ruled out as unsatisfactory (the « inept » set ), and those that are considered represent the « evoked » set , from which one alternative is likely to be purchased (Beatty and Smith, 1987).

The amount of effort a consumer puts into searching, depends upon a number of factors such as the market (how many competitors there are, and how great are the differences between brands expected to be?), item characteristics (how important is this product? How complex is the product? How obvious are the indications of quality?), consumer characteristics (how interested is a consumer, generally, in analyzing product characteristics and making the best possible deal?), and situational characteristics (Duncan and Olshavsky, 1982).

Decision-making (purchasing decision) process :

When evaluating alternatives, consumers choose from a list of acceptable alternatives (evoked set) based upon the criteria they have selected as being important. For a product to be considered by a consumer, he must know that it exists and perceive to be able to satisfy his needs. The criteria a consumer uses to choose between alternative items/services are the item/service attributes the consumer considers to be important. Consumers may make the purchase decision using compensatory or non-compensatory decision rules. Using a compensatory decision rule, the consumer identifies the important attributes, rates the alternative products on each attribute, and selects the product with the highest score. With a simple additive rule, the consumer selects the product that is judged to have the largest number of positive attributes. This is a relatively simple rule, used most often when motivation or ability is limited (Alba and Marmorstein, 1987). The weighted additive is a more complex compensatory rule in which the relative importance of each product attribute is also factored into the decision. Therefore, the consumer completes the more complicated task of computing a summated weighted score for each product on the salient attributes, and selects the product with the highest overall score. In contrast, non-compensatory decision rules do not balance all attributes and determine whether the positives outweigh the negatives. Rather, if the product does not meet a minimum standard on an important attribute, then it will not be considered. Using a conjunctive decision rule, the consumer sets minimum acceptable standards on all important attributes and eliminates any alternative that does not meet all the minimums. This helps consumers to narrow down the choices for further evaluation. If none of the products meet all the cutoff requirements, either the consumer must change the minimums acceptable, or change his or her decision rule (Grether and Wilde, 1984). With the lexicographic rule, the consumer first ranks the attributes in terms of perceived importance. Then, the alternatives are compared on this one most important attribute. If one scores sufficiently high on this most important attribute, then it is selected. If two or more are perceived as equally good, they are then compared on the second most important attribute. This process continues until the tie is broken. Consumers may use a combination of decision rules in choosing a product. First, they may use a rule to narrow down the choice set with some simple cutoff, and then they may apply a more complex compensatory rule to make the final choice. Some criteria are more salient than others, and those attributes will have a greater impact, or importance, in determining consumer selections.

Purchasing process  :

Through the evaluation process discussed above, consumers will reach their final purchasing decision, e.g. they go to the shop to buy the product/service. Purchase of the product/service can either be through the store, the web, or over the phone.

Post-Purchasing process (Satisfaction, dissatisfaction, cognitive dissonance)  :

Satisfaction/dissatisfaction : After the sale, the buyer will likely feel either satisfied or dissatisfied. If the buyer believes that s/he received more in the exchange than what was paid, s/he might feel satisfied . If s/he believes that s/he received less in the exchange than what was paid, then s/he might feel dissatisfied . Dissatisfied buyers are not likely to return as customers, and are not likely to send friends, relatives, and acquaintances.

Cognitive dissonance : Also called buyer's remorse. This post-purchase behavior is more likely to happen when the purchase is a more expensive one. The consumer may experience some regrets, or question himself/herself as to whether or not the purchase was a good one.

c. Consumption behavior models

- The models proposed by (Engel et al., 1968) (Howard and Sheth, 1969) and (Nicosa, 1966):

In order to understand decision-making related to human consumption, many researchers proposed several models for consumer behavior. The three main comprehensive models for this type of consumer decision-making, (Engel et al., 1968) (Howard and Sheth, 1969), and (Nicosa 1966), trace the psychological state and behavior of individual purchasers from the point at which they perceive a need, through the search for information, evaluation of alternatives, purchase, and final evaluation of the consequences. These models are presented in Fig 4.5, Fig 4.6, and Fig 4.7. The assumption in these models is that a purchase act is preceded by a sequence of mental information processing. This involves a cognitive function in forming beliefs, an emotional component in developing positive or negative attitudes, and a reaction, through which one is motivated to select and buy.

Although these three models help to understand the consumption behavior, they have specific drawbacks which are:

- They do not consider all the factors that influence consumption behavior. For example, they leave out the majority of external factors that come from the environment, which are presented in the following sub-section;

- They do not present, in detail, how each variable influences another variable, or a process. The relationships between the elements in the model (variables or processes) are presented by descriptive arrows;

- These models are very old and are not updated.

- The CONSUMAT model (Jager, 2000):

Recently, (Jager, 2000) and (Janssen and Jager, 1999) developed what the called ‘ the CONSUMAT model ’. It is a model of human behavior with the focus on consumer behavior. It combines in an elegant way many of leading psychological theories, such as theories about human needs, motivational processes, social comparison theory, social learning theory, theory of reasoned action, etc. The consumat model represents individual (or ‘ consumats ’) having needs which may be more or less satisfied, having items (‘ opportunities ’) to consume, and various abilities to consume these items. Futhermore, consumats have a certain degree of uncertainty. Depending on the combination of the indicators ‘ satisfied/not satisfied ’ and ‘ certain/uncertain ’, the consumats are engaged in four different cognitive processes: repetition, deliberation, imitation, and social comparison. When a consumat is both certain and satisfied, it has of course no reason to change its behavior, thus repetition is the strategy chosen. An uncertain but satisfied consumat has a reason to change its behavior. In this case the cognitive process chosen is imitiation of its neighbors. An unsatisfied but certain consumat on the other hand will deliberate. The final strategy is to consult the social network, the strategy chosen by uncertain and unsatisfied consumat (Janssen and Jager, 1999).

The Consumat model proposed by (Janssen and Jager, 1999) is relevant to our work in the sense it presents some variables or ‘indicators’ which can be integrated in our consumer model such as ‘needs’, ‘satisfaction’, ‘certainty’, etc. But, like the models proposed by (Engel et al., 1968) (Howard and Sheth, 1969), and (Nicosa 1966), the consumat model focuses on the human consumption behavior in general and neither focus on the spatial aspects of the consumption behavior nor on the geographic features of the simulated environment. Since we deal with geosimulation, these apects and features are critical to our work.

- The consumer model proposed by (Ben Said et al., 2001)

(Ben Said et al., 2001) presented a consumer model which was used in a simulation project called CUBES (Customers BEhavior Simulator). The objectives of this model are to develop a software simulating consumers’ behavior in a competitive market including several brands and to build a virtual population of consumers including several thousands of individuals, that reproduce real market properties (segmentation, evolution) independently of a given product. The CUBES model provides (1) the simulation of consumers’ behavioural attitudes (BA), (the impacts of consumption acts resulting from these attitudes), (3) retroactive effects of these acts on the consumers themselves, and (4) brand reactions to the market evolutions and their retroactive effects on the individual behavioural attitudes. The CUBES model covers the main concepts of previous theoretical works on consumer behavior such as consumer involvement, innovation diffusion and opinion leaders.

The consumer behavioural model presented by (Ben Said et al., 2001) is essentially based on a set of behavioral attitudes (BAs) defined basing on from social processes and personality traits. In this model the authors consider two social processes in order to model the interactions between the individuals of a virtual consumer population: Imitation_Process and Conditioning_Process . There are three BAs considered by the authors: Mistrust , Opportunism , Innovation . These BAs are defined based on consumers’ personality traits. In the CUBES model, the behavior of the consumer is based on these BAs.

The CUBES model of consumer presented by (Ben Said et al., 2001) is suited to design and develop an operational simulator of behaviors of thousands consumers in virtual markets. But, like the models proposed by (Engel et al., 1968) (Howard and Sheth, 1969), (Nicosa 1966), and (Jager, 2000) this model focuses on the consumption behavior in general and considers neither the spatial aspects of the consumption behavior nor the geographic features of the simulated environment.

The models discussed above provide a good basis to understand the elements which influence a consumer’s decision-making process. We extend them in order to propose an integrated model which includes the main factors influencing both, the consumption and the shopping behaviors, as well as the main processes composing them. This model is presented in Fig 4.8.

As shown in Fig 4.8, the initial version of the model contains several factors or variables that influence the shopping behavior (as mentioned in Section 4.3). In order to facilitate the reading of this model, some factors or variables are regrouped together in order to form groups of variables. For example, the shopper’s variables age and gender belong to the demographic profile , while the variables culture and social class belong to the social profile . Other variables are grouped together in order to form what we called ‘ states ’. For example, the variables time and money , that the shopper has to do shopping, belong to the possession state . The model also contains the main processes which are involved in the shopping behavior. For example, we can find the recognition needs process , the perception process , the alternatives evaluation process , the information search process , decision-making process , etc. (see Section 4.3).

This model will be used as a base: (1) to create the shopping behavior multiagent model (see Chapter 5 for the illustration of the third step of the method) and (2) to gather data concerning the shopping behavior using the survey (see Chapter 6 for the illustration of the fourth step of the method).

Due to the complexity of this model, we cannot take into account all these variables and processes in the shopping behavior geosimulation prototype. For example, in our model, we take into account the following variables: demographic (age, gender, occupation, sector of employments, marital status, etc.), interests, preferences, etc. and the main processes: Need recognition, perception, information search, alternatives evaluation, moving-decision, and moving processes.

The shopping behavior in a mall is quite similar to the shopping behavior inside a store, because these two behaviors are performed in geographic environments. In addition, the literature revealed that there exist several studies dealing with the shopping behavior in stores, which can be useful to study the shopping behavior in a mall. This similarity encouraged us to look at these studies in order to better understand the shopping behavior in a mall. In this sub-section we present the results of our study of human shopping behavior in stores. Based upon these studies, we can infer some characteristics of the shopping behavior in a mall. In addition, we focus only on the factors that come from the environment (the store) and that affect the shopping behavior, because the internal-external factors that influence the shopping behavior are the same factors that influence the consumption behavior and were mentioned in the previous sub-sections.

If we ask a shopper about his/her choice of an item in a store we can expect that he/she answers as follows: « It depends on when, where, why, etc. ». Consumer behavior depends enormously on the context or the situation. The influence of the situation on consumer preferences for a product or a service has been well documented in prior research on consumer behavior (Belk, 1974)(Srivastava et al., 1981). In line with (Belk, 1974), we define a usage situation as « those factors particular to a time and place of observation, which do not follow from personal (intra-individual) and stimulus (choice alternative) attributes, and which have a demonstrable and systematic effect on current behavior .» These factors are called stimuli or situational variables (Belk, 1974). Previous research such as that done by (Ratneshwar and Allan, 1991), has investigated the impact of different usage contexts on consumer consideration sets, and shown that consumers or shoppers consider different items in different usage situations. These findings are in line with those of (Warlop and Ratneshwar, 1993), who illustrate the importance of the usage context (familiar versus unfamiliar situations) regarding the formation of consideration sets.

(Belk, 1974) presented a situation as a part of a context, which is itself a part of an environment (see Fig 4.9). This distinction is very relevant because the situation is associated with a point in time and space.

In our work, we are interested in shopping situations. (Belk, 1974) presented shopping situations, based upon specific aspects, which are presented in the following points.

This aspect contains the following factors:

Visual factors : The visual factors of a physical environment such as a store that influences shopping behavior, are:

Lighting : Good lighting creates a pleasant atmosphere and encourages the shoppers to stay in the store longer. If customers go into a store that is dull and dingy, they are not encouraged to stay very long (Pellet 1996). However, if the lights are harsh and glaring, retailers are going to find that it takes a toll on employees’ performance. Glaring lights are also going to make the customers uncomfortable. If the light does not augment color contrast, it may not be suitable in a fashion wear store (Bandyopadhyay et al., 2001).

Colors and fixtures : Retailers frequently use colors and fixtures to improve the store environment. (Michon et al., 2005), given that they help provide a classy in-store ambiance. Proper color contrast can enrich the store image and perceived store or merchandise quality.

Layout and displays : It is an established fact that changes in layout can affect the sales of a store (Bandyopadhyay et al., 2001). These authors’ research on the impact of layout and display effects on sales revealed several interesting findings.

Non-visual factors : These factors are

Sound and music : Sound plays an increasingly critical role in helping retailers to entertain and inform shoppers (Michon et al., 2005). Retailers are now expanding its use beyond the traditional background music application. Sound helps to create a unique environment as merchandisers are confronted more and more with the monotony of merchandise in most stores (Michon et al., 2005). Sound, in the form of music, is commonly used to entertain shoppers. It can also give a significant boost to a store’s energy level by creating a sense of excitement. The audio experience, therefore, can be an important component of a store’s overall entertainment factor. The main advantage of using sound in the form of music is that it can greatly extend customers’ stay time. Music is known to have a classical conditioning effect on consumers (Bandyopadhyay et al., 2001). If the music is likeable, customers not only feel good, but they tend to stay longer likewise. Music also makes the service environment look more positive to customers in waiting lines (Bandyopadhyay et al., 2001).

Odors and smell : Studies have shown that ambient aromas both influence a consumer’s mood and the time he/she spends in the store. What’s more, the sense of smell has the greatest impact on emotions because, anatomically speaking, the nose is directly connected with the olfactory lobe in the limbic system of the brain, which controls emotion (Maclean, 1973).

Much research examined the factors that affect shopping behavior in retail environments. Of existing studies, most have discussed the environmental factors, or physical surroundings, relative to bank, travel agency, hotel, restaurant, hospital, and workplace settings. Table 4.2 shows some of the environmental factors that have been discussed in the literature (Baker et al., 1988), (Bitner, 1992), (Engel et al., 1995) and (Lewison, 1994). (Baker et al., 1988) studied banks, (Bitner, 1992) discussed a typology based upon who performs actions within the environment, as well as having proposed a conceptual framework for three types of service organizations: self-service, interpersonal service, remote service. (Engel et al., 1995) and (Lewison, 1994) provided textbook discussion on the environment factors.

The time has played a major function in several research disciplines that are closely related to consumer behavior. This dimension remarkably affects consumption and shopping behavior (Carmon, 1991) and (Dhar and Nowlis, 1999). The importance of time, as a major variable of interest in consumer behavior theory, had already been recognized in the early stages of consumer research. Over the past twenty years, several research streams concerning time have evolved within the consumer behavior literature. These included the effects of time pressure on consumer’ decision-making, allocation of their time, and perception of time. Several interdisciplinary reviews also appeared in the marketing literature (Carmon, 1991). The majority of research works addressed the effect of time pressure when consumers are forced to choose. Some researchers such as (Dhar and Nowlis, 1999), examined the decision process and choice outcomes when the no-choice, or deferral option, is available. Studying the no-choice option in consumption or shopping behavior is important for our study because mall’s visitors are not always obliged to buy items or to visit stores. (Dhar and Nowlis, 1999) presented a model of the consumer decision-making process (alternative evaluation process) under time pressure. This model is presented in Fig 4.10.

These models offer the following options under time pressure:

Forced choice : The consumer does not have the choice to buy the item. In the forced choice, we have two issues:

The consumer can have a conflict : If there is a conflict (i.e., when the choice is more difficult and the alternatives are similar in terms of importance to the consumer; all the alternatives are attractive), he-she can change the strategy in his-her alternatives’ evaluation process. For example, and in order to make a choice, the consumer can change the evaluation of the alternatives’ attributes in order to remove the conflict, and he/she buys his-her item because he/she does not have a choice.

There is no conflict : The consumer buys his-her item because he-she does not have any choice.

Deferral choice : If they have the choice, consumers are more likely to select the no-choice option when conflict is high (when all the alternatives are attractive) rather than when conflict is low (there is a single superior alternative). An analysis of the decision-making process suggests that consumers who expressed more thoughts, or made more comparisons (and presumably found the choice more difficult), were more likely to choose the deferral option. An implicit assumption that underlies such a pattern of preferences is the notion that selecting the best option, within the choice set, precedes the deferral decision. Thus, the difficulty of making the selection when choice among the alternatives involves conflict, increases the tendency to defer choice.

Based upon the results of studies presented in the previous sub-sections, dealing with consumption and shopping behaviors in stores, we can list the main characteristics of the individual shopping behavior in a mall.

Shopping behavior in a mall has many dimensions:

■ It is a personal behavior : It is influenced by personal variables such as demographic variables (gender, age, occupation, marital status, etc), or the position in the family or in a group (who makes the decision.), and other personal characteristics of the individual.

■ It is a psychological behavior : It is influenced by psychological factors such as motivations, attitudes, values, personality (or self-concept), lifestyle, mood, emotion, involvement, etc.;

■ It is a cognitive behavior : It is influenced by the perception, memorization, and knowledge of the person about his/her environment;

■ It is a social behavior : It is influenced by the leading opinion in the family or group, the reference group, the role in the family or group, the social class, the culture, the sub-culture, the religion, etc.; and

■ It is an environmental/situational behavior : It is influenced by three aspects of the environment:

□ Spatial aspect: It is spatial behavior because it is influenced by the spatial and geographic characteristics of the environment (mall) such as the layout, the architecture of the environment, the colors and textures, etc.;

□ Temporal aspect: It is influenced by the temporal factor; and

□ Atmospheric aspect: It is influenced by the atmosphere of the environment: the music, the odor, the temperature, the humidity, the lighting, etc.

All these factors can be categorized by two classes: Internal or external factors.

The shopping behavior in a mall is composed of the following processes:

■ The needs/problems recognition process : The consumer comes to the mall with the need to visit certain stores in order to purchase specific items or services;

■ The information search process (perception and memorization processes): The shopper perceives the stores, kiosks, etc.;

■ The alternatives’ evaluation process : The shopper can have the choice of visiting many stores or kiosks to satisfy his/her needs, so he/she must decide which stores/kiosks to visit;

■ The decision-making process : The customer decides which store to visit;

■ The acting process : The customer acts by moving to his/her selected store; and

■ The post-decision process : The customer leaves the stores, and after his/her shopping trip, he/she leaves the mall.

Based upon the study made in several disciplines, we designed a theoretical model of shopping behavior in a mall. This model contains internal and external factors that affect shopping behavior and processes that compose it. This model is presented in Fig 4.11.

The shopping behavior can be accomplished in groups which can be families, friends, colleagues, etc. The social aspect of shopping behavior brings out the idea of studying the shopping behavior of groups, in addition to the studies done with regards to individual behavior. In order to understand the shopping behavior of groups, we turned towards the discipline of consumer behavior. Unfortunately, the majority of research that involves group consumption behavior deals with one category of group which is the family . Moreover, these studies focused on the investigation of family’s influence on the decision-making process of one consumer belonging to the family.

Due to the noteworthy absence of studies dealing with the consumption or shopping behavior of groups we turned towards other disciplines that studied groups and their behaviors. Among these disciplines, we can cite: sociology, social psychology, and the discipline that studies group decision-making. In Section 4.4.1, we present respectively, important results of our studies that can help us to better understand the groups, their structure, and their behavior. Then, in Section 4.4.2, we present how we can specify the group within the scope of shopping behavior in a mall.

In sociology, a group is usually defined as ‘ a collection consisting of a number of people who share certain aspects, interact with one another, accept rights and obligations as members of the group, and share a common identity ’ (Hare, 1976). While an aggregate comprises merely a number of people, a group in sociology exhibits cohesiveness to a larger degree. Aspects that members in the group may share include interests, goals, values, ethnic/linguistic background, and kindship. To define a group, we present the following key points:

■ Members have ‘ self-definition ’ as a group member;

■ Members are identified by others (‘ other-definition ’) as a group member. It is especially important that other group members also identify the individual as belonging to the group. Both ‘ self ’ and ‘other’ definition as a member are necessary. Without self-definition, the individual will lack motivation to act in the group’s best interests, and may even think it silly or futile to conform with group norms. Individuals who are compelled to be in a group lacking self-definition may even act to undermine the group. The person may require surveillance to stay in the situation (as for example in a prison). At the very least, someone may enjoy group benefits without contributing (freeloaders) because s/he does not self-define as a group member. Without other-definition, we may perceive the person as a fraud and this could expel him/her from the group. At the very least, we may create distance mechanisms (such as ignoring the person, «rewriting history» to exclude him/her from the group, or literally, «cutting them out of the picture»). Someone who is «in, but not of» the group is not seen as entitled to group privileges or rewards.

■ Interdependent and common goals that require membership coordination: This characteristic implies commonality in the group. Interdependence means that people cannot achieve goals individually, but must do so as group members. Interdependent goals require coordination among the members to achieve them. They require the coordinated efforts of at least two people working together. Interdependent goals imply a division of labour so that each member has a unique and specific task i.e., a specialized task.

■ A common fate: This characteristic implies commonality in the group.

■ Direct interaction among members: Ironically, this characteristic typically comes to mind first.

In the literature, we found several types of groups. For example, theorists often distinguish between primary groups and secondary groups . The primary group consists of a small group with intimate, kin-based relationships; families, for example. This group is important for the individuals who compose it. In contrast to the primary group, the secondary group is large, whose relationships are formal and institutional. This group is less central for the individuals who compose it (Bishop and Myers, 1974).

The group can be formal or informal . An informal group is characterized by interaction that is more informal: diffuse, socio-emotional, and particularistic. The informal group dissolves when all the original members leave. While a formal group is characterized by formal interaction. This group survives at least one complete turnover of members. Formal groups often have more resources and advantages than informal groups. For example, a formal group may be more likely to have a group culture containing:

■ Codification or some form of recorded information about the group such as minutes or a ledger.

■ History, such as a group biography, or an oral tradition of stories or accounts.

■ Group symbols, such as a flag, mascots, flowers, or emblems.

■ A special group language or set of phrases known only to group members.

■ Group rituals, such as a special handshake, or a protocol order for meetings.

A group can include membership or non-membership. We can conceptualize about being a group member or not. One can be a member of a group with very sharp boundaries, such as a classroom. Other group memberships, such as study groups or friendship groups, are more fluid. You come and go at your own convenience, sometimes, defining yourself as a member, and sometimes not.

A group can be ascribed or achieved . The membership of an ascribed group usually consists of individuals who either are born into the group, marry into the group, or who are otherwise placed there, a priori, by the norms and customs of a culture or subculture. Many important primary groups, such as families or religious affiliations, are ascribed groups. Members in an achieved group must accomplish a task, or demonstrate a skill, in order to enter the group. Recruits to youth gangs, or fraternity and sorority pledges, typically must complete a pledgeship or initiation to show that they deserve membership.

Social roles emerge from groups. In fact, in nearly all cases, the division of labor or tasks in groups is the major cause of creating social roles. Even the most rudimentary groups tend to create a specialized division of labor and interlocking positions.

As groups survive, they become more complex, taking on many new tasks. Such a proliferation of tasks creates a coordinated and interdependent division of labor. The group can accomplish tasks that no one member could achieve alone. As a group grows in size, it also becomes easier to create a specialized division of labor. The increase in people available to do ‘group work’ means that we can pick individuals with the appropriate talents, skills, and interests for each job. Thus, from this division of labor frequently comes the development of specialized social roles. A role is a social position with an accompanying, attached set of rights , duties , and scripts (Bishop and Myers, 1974). When roles are arranged in some type of hierarchy, or ‘chain of command’, we have a stratified social system (Bishop and Myers, 1974). The most critical aspect of roles is that they transcend individuals because they are social positions: anyone who occupies a specific role is expected to display a minimal level of its scripts and duties. Some role requirements are codified and formal, while others are informal. While latitude often exists in how to play a role, there are also usually minimal defining criteria (Cao and Frada, 1999).

Group decision-making has long been a major topic in a wide variety of social sciences: anthropology, sociology, political science, marketing, and psychology (Hare, 1976).The basic thrust of this research has been to determine those factors that affect the process, by which groups make decisions (Swap, 1983). In consumer research, groups that have been studied include families (Davis, 1976) and organizational buying centers (Webster and Wind, 1972). Research has been centered upon identifying the individuals involved in the decision-making process (Davis and Rigaux, 1974), (Silk and Kalwani, 1982), decision role structure (Davis 1976), (McMillan, 1973), and the determinants of relevant influence (Kriewall, 1980), (Thomas, 1982).

In consumer research (or the social sciences in general), limited progress has been made with respect to developing and testing mathematical representations of group processes that can then be used to predict decisional outcomes in consumer behavior. Three exceptions in marketing and consumer research include: (Choffray and Lilien, 1980), (Corfman and Lehmann, 1987), and (Eliashberg and al., 1986).

■ (Choffray and Lilien, 1980) proposed a set of four models that mathematically transform a set of individual choice probabilities into a group choice probability. Each model corresponds to a different conceptualization of the interaction process within the group, which, whether or not members vote, uses the inputs of individuals in proportion to their importance, search for consensus, or attempt to be least perturbing to other members. Unfortunately, the authors did not test these models empirically.

■ (Corfman and Lehmann, 1987) developed and tested a different set of algebraic models, in which the group members’ personal traits (i.e., resources and goals) are used to predict the outcomes of conflict resolutions. The forms of the models are basically linear, or linear with interactions. Models are estimated using data collected on couples’ decisions. The dependant variable is the probability of getting one’s own way. The independent variables are the resources that an individual brings to the group decision (e.g., expertise, social debt), relative to those of his or her partner(s). While this initial work is a step in the right direction, the particular algebraic forms seem ad hoc and not well-grounded in psychological theory.

■ (Eliashberg et al., 1986) tested models from the decision-analysis literature, which follow from sets of axioms, to predict a group’s preference judgment as a function of the judgments of its individual members.

(Rao and Steckel, 1991) proposed an interesting model called the ‘ polarization model ’. This model is used to infer the objective of a group from those of the members that compose it.

Based upon the study results presented in the previous sub-section, we can specify the characteristics of a group of shoppers that come to shop in a mall. Like an individual shopper we propose to characterize a group of shoppers by a structure and a behavior.

In this chapter, we presented the two first steps of our proposed method. The first step aims to identify the application domain of the geosimulation, its future end-users, as well as the needs of these users. In the second step we identify the characteristics of the phenomenon to be simulated and those of its environment.

The chapter also presented an illustration of these two steps using the customer’s shopping behavior as a case study. In the illustration of the first step, we defined the main users of the shopping behavior geosimulation which are mainly mall managers. These managers need to use the geosimulation as a decision-making tool that helps them to make decision about the configuration of their mall in order to make it more comfortable for the shoppers.

In the literature, we found little research dealing with the shopping behavior in a mall. For example, (Dijkstra et al., 2001) and (Timmermans et al., 2003) simulated the shopping behavior in a mall as a pedestrian behavior. However, they did not study the factors that influence this behavior, the processes that compose this behavior, etc. (Ruiz et al., 2004) examined the shopping behavior and collected relevant data, but they contented themselves with an analysis of this data in order to define the patterns of shopping behavior in a mall. In their study, they omit an important factor, which is the spatial factor that influences the shopping behavior in a mall. Due to this lack of research concerning the shopping behavior within a mall, we turned to other disciplines in order to study such behavior, or similar conduct, such as human consumption behavior. Based upon this study, we were able to specify the shopping behavior in a mall. This specification contains the majority of relevant factors that influence such a conduct, the main processes that compose this behavior, the models that describe it, etc. In our study and specification of the shopping behavior, we are not limited to the individual aspect of this behavior, and we studied the aforementioned with regards to groups. To do that, we turned to the discipline of social sciences in order to investigate a group’s decision-making process. Based upon our research, we can assert that there has been no study, nor simulation of a group’s shopping behavior in a mall.

Based on all the shopping behavior’s studies results which are carried out in several disciplines, we developed an initial version of the shopping behavior model. This model contains the majority of factors influencing the shopping behavior in a mall, as well as the processes that compose it. This model is used as a basis in the next steps of the method. The majority of models proposed in the literature and describing the shopping behavior ((Engel et al., 1968), (Howard and Sheth, 1969), (Nicosa, 1966), etc.), did not take into account the spatial characteristics of the shopping behavior, nor the geographic features of the mall. Since we deal with geosimulation, these characteristics are important to us. For this reason, we integrate them in our proposed shopping behavior model. Presenting such a model can be considered as a contribution in the consumer behavior field and the fields simulating the shopping behavior.

In the next chapter we present the third step of our proposed method in which we design the multiagent geosimulation models. The design of these models is based upon the results of the first two steps.

© Walid Ali, 2006