An Effective Customization Procedure with Configurable Standard Models

Publication Type

Journal Article

Publication Date

2005

Abstract

In electronic catalogs, commodities such as computers and electronic equipment are specified as standard models although a variety of possible alternative specifications can exist as a combination of selected options; therefore, customized configurations are essential to support various customers with individual needs. Thus, the problem here is the selection of a standard model and reconfiguration with this selected model. An issue is that requirements may be fulfilled by more than one standard model. To develop an algorithm that can find the near minimum price without causing unacceptable computation effort, we devised the Standard Model Selection and Modification (SMSM) Algorithm. To establish the SMSM Algorithm, we propose the Concurrent Local Propagation procedure complemented with pruning capability owing to the nature of standard models. The effective strategies for selection of seed variables and stopping rules are devised through comparative experiments. For the experiment, we use Dell's personal computer (PC) products consisting of 42 standard models with 25 specification variables. The SMSM Algorithm is tested with 75 experimental cases, and we found that the most similar standard models could discover the minimum price only in 24 out of 75 cases and that the SMSM procedure could reduce the price by 6.04% from the one offered by the most similar standard models.

Keywords

Configuration, Customization, Case-based reasoning, Comparison shopping, Constraint satisfaction problem

Discipline

Computer Sciences | E-Commerce

Research Areas

Information Systems and Management

Publication

Decision Support Systems

Volume

41

Issue

1

First Page

262

Last Page

278

ISSN

0167-9236

Identifier

10.1016/j.dss.2004.06.010

Publisher

Elsevier

Additional URL

http://dx.doi.org/10.1016/j.dss.2004.06.010

Link to Full Text

Share

COinS