A data-driven approach for the optimisation of product specifications. Issue 3 (1st February 2019)

• Record Type: Journal Article
• Authors: Zhang, Lei; Chu, Xuening; Chen, Hansi; Yan, Bo
• Is Part Of: International journal of production research. Volume 57:Issue 3(2019)
• Page Start: 703
• Page End: 721
• Publication Date: 2019-02-01
• Subjects: operating data -- product optimisation -- conjoint analysis -- customer choice model -- customer satisfaction function; Factory management -- Periodicals; 658.57
• DOI: 10.1080/00207543.2018.1480843
• Languages: English
• ISSNs: 0020-7543

Abstract

Abstract : In order to develop the profit-maximising, market share-maximising or cost-minimising bundle of product engineering specifications with proper performance levels, an optimisation model driven by operating data is proposed. The operating data are input as the sources to conduct the optimisation and a data-based customer satisfaction function can be formed. Then, a customer choice model developed from the customer satisfaction is constructed to estimate the customer choice probability. The expected market share (EMS) then can be derived from the choice probability. After all, a multi-objective model is constructed to maximise the EMS and minimise the total engineering cost. The candidate Pareto-optimal solutions can be obtained by solving the optimisation model. Then a membership function is defined to select the optimal solution from the Pareto-optimal solutions. A case study for optimising the smartphone's specifications is conducted to demonstrate the effectiveness of the newly developed approach. Compared with the commonly used Conjoint Analysis (CA) method in determining the most desired levels for product specifications, the proposed data-driven method can avoid the situation where the user's preferences are irrational, making the proposed method be more practical in measuring customer preferences than the utility-based model.