Developing a marketing decision model using a ... - Semantic Scholar

3 downloads 97276 Views 280KB Size Report
Developing a marketing decision model using a knowledge-based system. U. Yavuz a ..... product life cycle, was implemented in an automated knowledge base.
Knowledge-Based Systems 18 (2005) 125–129 www.elsevier.com/locate/knosys

Developing a marketing decision model using a knowledge-based system U. Yavuza, A.S. Hasiloglub,*, M.D. Kayac, R. Karcioglud, S. Ersoze b

a Communication Faculty, Ataturk University, 25240 Erzurum, Turkey Department of Electronics and Telecommunications Engineering, Ataturk University, 25240 Erzurum, Turkey c Vocational College of Erzurum, Ataturk University, 25240 Erzurum, Turkey d Faculty of Economics and Administrations, Ataturk University, 25240 Erzurum, Turkey e Vocational College of Kirsehir, Gazi University, Kirsehir, Turkey

Received 18 October 2003; accepted 1 December 2004

Abstract This paper describes using a knowledge-based system for developing a marketing decision model. The approach used in this study uses a decision table as a knowledge engineering tool. The decision table is used as a means of representing a set of decision rules to construct a developed marketing decision model. To support the modeling process, Prologa, an existing decision table engineering workbench, is used. The developed marketing decision model is used to determine the entrance time of a new product into market by utilizing knowledge-based systems. Presentation of a new product to the market at the best time will provide an advantage to competing companies and will increase their market share. q 2005 Elsevier B.V. All rights reserved. Keywords: Decision tables; Knowledge based systems; Product life cycle; Prologa

1. Introduction Product life cycle is one of the oldest concepts in analyzing and solving business problems [1]. Life cycle refers to the period from the product’s first launch into the market until its final withdrawal. Although life cycle varies by product and sector base, usually there are usually four phases in the life cycle period as shown in Fig. 1. The first period is the entrance phase, the second period is the development phase, the third period is the maturity phase and the fourth period is the satisfaction phase. The entrance phase is the period when a product is introduced to the market and an effort is made for its acceptance. In general, this is the period of catching up at par point. The development phase is the best step; the product has been through the brightest period and reached its maximum profit. In the maturity phase, problems

* Corresponding author. Tel.: C90 442 231 48 73; fax: C90 442 236 09 57. E-mail address: [email protected] (A.S. Hasiloglu). 0950-7051/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.knosys.2004.12.002

gradually occur and sales start to decrease. Despite this decrease, companies try to keep sales high by using other marketing techniques. Which are called other sales efforts? In that period increases in sales like jumping sales (comb tooth) occur. The satisfaction phase is the period that the companies least prefer to be in because they will start to lose in a while. Companies struggle to survive in competition by designing new products instead of products which have a decreasing product life cycle. For this reason, it is important for companies to determine the life cycle of a product. The purpose of this study is to investigate the entrance time of a new product into the market by utilizing knowledge-based systems. The decision table (DT) is used as a technique to model the diagnostic and strategic knowledge involved. This approach uses the DT as a knowledge engineering tool, and incorporates a DT analyzer that shows rule conditions, which can be used to query experts. The DT analyzer also reports redundant rules and provides assistance for developing rules that use as few variables as possible. This has the benefit of limiting the number of rules in the knowledge base [5]. In addition, in the DT approach, validation and verification (V and V) of the knowledge base can be done in the modeling phase instead of being performed afterwards

126

U. Yavuz et al. / Knowledge-Based Systems 18 (2005) 125–129

Fig. 1. Life cycle period of a new product.

on the implemented systems and it was reported earlier that, in a vast majority of cases, the DT technique is able to provide for extensive validation and verification assistance [3]. Performing V and V during the early development phase strongly improves the process of knowledge acquisition and representation and prevents errors that would be more expensive to fix at a later time [2,4,7]. The model is applied in a knowledge-based system (KBS) where it can be used in combination with models of global market, target market, manufacture and performance. The rest of this paper is organized as follows. In Section 2 the DT representation is briefly described. Section 3 analyzes which factors affect product life cycle periods. In Section 4 (the main part of this paper), the major modeling issues of the study are examined, based on the correspondence between a DT and a knowledge based system. The paper concludes with a summary of the findings and directions of future research.

2. Factors affecting product life cycle period Presentation of a product to the market is caused by the opinion of innovation or requirements. The criteria of existence of a product in the market are listed below: 2.1. Effect of economical changes Economical decisions, made by the government and economical structure formation in the market, are affective on product life cycle period. The economic conditions mentioned above cause constriction or expansion. Constriction causes crises and shock, expansion causes gap in the market. Changes in the economical structure affect the introduction of a product to the market and, the withdrawal

of a product from the market; therefore, it is a major factor affecting the product life cycle period (see Fig. 2(a)). 2.2. Effect of political, social and cultural structure changes Companies have to take into consideration the political and cultural characteristics of the individuals and the opinions of society, which are formed by those individuals in a region called market in which the consumer and the companies get together. Political, social and cultural changes that occur in society will change market conditions (see Fig. 2(a)). 2.3. Effect of quality Quality and cost are the two main factors affecting consumer preferences and, therefore, product life cycle period and the relationship between them is very complicated. As a common opinion in literature quality and cost are directly proportional. Higher quality in production increases costs. Unqualified production means doing a job more that once and wasting machine hours, labor hours and raw material. 2.4. Effect of consumer preferences Life cycle period is affected by consumer preferences and social–cultural structure changes. Consumer social classes are in a very close relationship with family structure, leader groups, and reference groups. Social rules affect and are affected by these groups. Other than those affects mentioned the consumer’s physical structure; learning perception, motivation, personality, attitude, belief, etc are the factors, which cause consumers to make internal reactionary behaviors, which are unpredictable. Quality

U. Yavuz et al. / Knowledge-Based Systems 18 (2005) 125–129

127

Fig. 2. Decision table representation.

and consumer preferences are the factors that define the manufacture point (see Fig. 2(b)).

products new techniques of technology treatment new production methods and new organization structures (see Fig. 2(b)).

2.5. Effect of technological changes 2.6. Effect of competition Technology is the knowledge, which is used or can be used for production of goods and services and capability of production and use of that knowledge. In other words, a build up of knowledge converts inputs into meaningful outputs. It is not always possible to acquire development and transfer of technology easily and cheaply. But successful organizations are always more powerful in technology management than unsuccessful ones. Technological capability can be the only or main reason of increase or decrease of major market share between competitors. Technological renewal has brought about new

The main principle of competition is to be powerful and to survive. Current products are always under the threat of new products introduced to the market in every stage of their life cycle period. Competition which is done by means of information in our present day is an individual’s or group’s desire to race for the same purpose and to excel each other. From a company’s point of view, competition is the entire activity of the companies trying to provide goods or services to the market directly or indirectly (see Fig. 2(c)).

128

U. Yavuz et al. / Knowledge-Based Systems 18 (2005) 125–129

2.7. Effect of other selling efforts As mentioned above, in the maturity phase problems arise gradually and sales begin to decrease. Despite this sales decrease, companies try to keep sales high by using other marketing activities, which are called other sales efforts. If other selling efforts show an increase, it can be said that the sale of the manufacture is decreased (see Fig. 2(c)). It is plain to see from a review of the conventional life cycle that profit has started to fall in spite of the proportional increase in sales.

3. Structure of the model The aim of the proposed system is to model the concepts and rules of marketing decisions. In this chapter, a set of DTs will be constructed modeling the domain knowledge (see Fig. 3). Prologa is used to build up the modeling process. Prologa is an interactive design tool for computersupported management and the construction of DTs. In the conventional product life cycle, the introduction of a new product to the market corresponds to point A in Fig. 1. When a company comes to this point at the end of the maturity period, it has to choose the alternative of either new product, new market or withdrawal of goods from the market, so as not to enter into the fourth period, regression. Depending on the company’s structure the new product alternative can be a new product in the physical/functional context, a new product in the consumer’s view or an alternative use of the product. Point A in Fig. 1 in the existing systems is considered to be late for a new product to enter the market, because, this point is when the company withstands a number of other costly sales efforts (promotion, excess goods, discount, etc) to keep the sales active. It is plain to see from reviewing the conventional life cycle that profit has started to fall in spite of the increase in sales.

It is suggested in the proposed system to determine the point specified as point A in Fig. 1 using the expert system. In this system, point A can be taken to an earlier time in the existing policies. In operating the system, product life cycle maturity period characteristics will be reviewed and efforts will be made to determine the most suitable time for presentation of the product to the market by evaluating of the factors named as macro and micro market indicators. In this study, 33 rules have been determined for evaluation of macro and micro market indicators and maturity characteristics. These rules have been transferred into DTs using of Prologa. The overall economic situation and legal and political circumstances prevailing in the market factors, named as global market indicators are reviewed. The target market indicator factors are a renewal of the product and manufacture points. The manufacture point is compares of performance of the product and its rivals, as shown in Fig. 2(b). The result of the review reveals the probability that performance of the product can be lower than, higher than or equal to that of the closest rival product. In this DT, the condition of the ‘manufacture point’ has three condition domain factors: m.p.!c.m.p., m.p.Zc.m.p., and m.p.Oc.m.p. Above, ‘m.p.’ is the manufacture point of our product; ‘c.m.p.’ is the manufacture point of our competitor’s product. Indicators of manufacture are competition, other selling efforts and proportional increase in sales as shown in Fig. 2(c). The most striking signs of the maturity stage are a decrease in competition, rivals’ introduction of new products to different market portions and constriction of the existing market. As a result of sales rate decrease, the company will initiate other sales efforts to increase sales. These efforts will escalate the cost of other sales efforts. Thus, the profit rate will drop because a big portion of the profit is used to finance other sales efforts. Indicators in Fig. 2(d) for the maturity characteristics in product life cycle were reviewed. DTs of global market, target market, and manufacture were combined and the DT of performance was obtained. As a result, the decision rules of the system are constructed as following: If PERFORMANCEZActive Then ‘Preserve the present status’ End if If PERFORMANCEZPassive Then ‘Introduce new product to market’ End if

Fig. 3. Example of a system-user dialogue:product life cycle.

If PERFORMANCEZBad Then ‘Withdraw the product from market’ End if

U. Yavuz et al. / Knowledge-Based Systems 18 (2005) 125–129

4. Results The DT based marketing decision model, which defines product life cycle, was implemented in an automated knowledge base. We used Prologa as an expert system development tool for product life cycle. As a result of operating the expert system, three different deductions can be made: preservation of the present status, introduction of a new product to the market and withdrawal of a product from the market. Fig. 3 shows an example of system-user dialogue for product life cycle. As a result of the consultation, the action reached was ‘bad’. Therefore, the decision rule for this consultation is ‘withdraw the product from the market’.

129

Values taken by these reviewed factors are interpreted by means of the expert system and the best decision for the company is made. As a result of the study, the most suitable time for introducing a product to the market was determined. This structure can be conveniently used for different sectors with new rule bases to be obtained from experts of the other sectors. In further research, to get a most realistic model, it is possible to add quantitative parameters to model such as production per unit time, wasting machine hours, labor hours, and raw material. Besides, to adapting this model to real life, general rules should be extended. For example, to define political circumstances of the target society, new rules set can be added to model.

5. Summary and conclusions References The DT technique was used to represent expert knowledge in terms of a set of DTs. Each DT represents in each problem area an exhaustive and exclusive set of decision rules to identify problems, analyze problems and formulate actions. Structuring the decision problem is the most important function of knowledge-based systems. In the context of knowledge-based systems, DTs are used efficiently and effectively during the verification and validation process, and in knowledge acquisition [6]. Many expert systems built now-a-days are prepositional expert systems, equivalent to DTs. A procedure to perform the transformation between prepositional expert systems and DTs (and thereby substantially increase execution speed) is presented in [2]. In this study, we aimed to carry out a determination of the introduction of a new product to the market and review the product’s life cycle maturity period characteristics as well as the factors called macro and micro market indicators.

[1] H. Bauer, M. Fischer, Product life cycle patterns for pharmaceuticals and their impact on R and D profitability of late mover products, Int. Business Rev. 9 (2000) 703–725. [2] R. Colomb, C. Chung, Very fast decision table execution of prepositional expert systems, In: Proceedings of the 8th National Conference on Artifical Intelligence, AAAI90, (AAAI Press/The MIT Press, Boston, Massachusetts, 1990) 671–676. [3] B. Cragun, H. Steudel, A decision-table based processor for checking completeness and consistency in rule-based expert systems, Int. J. Man–Machine Stud. 5 (1987) 633–648. [4] J. Grant, J. Minker, The impact of logic programming on databases, Comm. Offhe ACM 35 (3) (1992) 66–81. [5] J.P. Seagle, P. Duchessi, Acquiring expert rules with the aid of decision tables, Eur. J. Oper. Res. 84 (1995) 150–162. [6] J. Vanthienen, E. Dries, Illustration of a decision table tool for specifying and implementing knowledge based systems, Int. J. Artif. Intell. Tools 10 (1994) 267–288. [7] J. Vanthienen, G. Wets, An illustration of verification and validation in the modelling phase of KBS development, Data Knowledge Eng. 27 (1998) 337–352.

Suggest Documents