A Multi-Attribute Decision-Making Model for Construction Enterprise Resource Planning System Selection. Issue 1 (2nd January 2016)

• Record Type: Journal Article
• Authors: El-Mashaleh, Mohammad S.; Hyari, Khaled H.; Bdour, Ahmad N.; Rababeh, Shaher M.
• Is Part Of: International journal of construction education and research. Volume 12:Issue 1(2016)
• Page Start: 66
• Page End: 79
• Publication Date: 2016-01-02
• Subjects: Data envelopment analysis -- decision making -- decision support system -- enterprise resource planning -- information technology; Building -- Study and teaching -- Periodicals; Construction industry -- Study and teaching -- Periodicals; 624.07
• DOI: 10.1080/15578771.2015.1015755
• Languages: English
• ISSNs: 1557-8771

Abstract

ABSTRACT: Enterprise resource planning (ERP) systems are increasingly becoming essential business technologies. Selection of the optimal ERP system among numerous alternatives is regarded as a major challenge for the decision-maker. This choice is complicated by multiple attributes that the decision-maker needs to consider when evaluating different ERP options. The purpose of this article is to propose a multi-attribute decision-making model for ERP system selection based on data envelopment analysis (DEA). The strengths of the proposed DEA model include the input-output framework, which allows decision-makers to incorporate multiple attributes when evaluating different ERP options. These attributes are not required to be congruent. They can be measured by such criteria as monetary sums, time, or subjective score. Additionally, DEA can accommodate a large number of options (i.e., ERP systems) and variables (i.e., attributes). The proposed approach was utilized by a construction contractor to select the optimal ERP option out of eleven alternatives. During the evaluation process, the decision-maker incorporated six attributes to make the decision. These considered attributes are total cost, implementation schedule, functionality, user friendliness, customization capability, and service and support quality. Sensitivity analysis proved that the proposed model is stable and robust.