An Effective Customization Procedure with Configurable Standard Models
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.
Configuration, Customization, Case-based reasoning, Comparison shopping, Constraint satisfaction problem
Computer Sciences | E-Commerce
Information Systems and Management
Decision Support Systems
LEE, Hyun Jung and LEE, Jae Kyu.
An Effective Customization Procedure with Configurable Standard Models. (2005). Decision Support Systems. 41, (1), 262-278. Research Collection School Of Information Systems.
Available at: http://ink.library.smu.edu.sg/sis_research/1182