Development of a hybrid methodology for ERP system selection
Published on Oct 1, 2014
· DOI :10.1016/j.dss.2014.06.011
Enterprise resource planning (ERP) systems that aim to integrate, synchronize and centralize organizational data are generally regarded as a vital tool for companies to be successful in the rapidly changing global marketplace. Due to its high acquisition-purchasing, installation and implementation-cost and the wide range of offerings, the selection of ERP systems is a strategically important and difficult decision. Since there is a wide range of tangible and intangible criteria to be considered, it is often defined as a multi-criteria decision making problem. To overcome the challenges imposed by the multifaceted nature of the problem, herein a three-stage hybrid methodology is proposed. The process starts with the identification of most prevailing criteria through a series of brainstorming sessions that include people from different organizational units. Then, due to the varying importance of the criteria, a fuzzy Analytic Hierarchy Process, which handles the vagueness inherent in the decision making process, is used to obtain the relative importance/weights of the criteria. These weighted criteria are then used as input to the Technique for Order Preference by Similarity to Ideal Solution method to rank the decision alternatives. As a real-world illustrative case, the proposed methodology is applied to the ERP selection problem at Turkish Airlines. Because of the collaborative and systematic nature of the methodology, the results obtained from the process were found to be highly satisfactory and trustworthy by the decision makers. ERP system selection is a complex multi-criteria decision making process.Imprecision and large-number of criteria are among the most important challenges.Use of a hybrid methodology improves the decision outcomes.Fuzzy logic, AHP and TOPSIS are collectively and synergistically used in this study.The proposed methodology produced encouraging results to the ERP selection problem.
Figures & Tables