Analysis of Online Marketplace user Satisfaction using SERVQUAL, Quality Function Deployment, and Mixed Integer Linear Programming

Nayla Nur Alifah (School of Data Science, Mathematics, and Informatics, IPB University, Bogor, 16680, Indonesia)
Toni Bakhtiar (School of Data Science, Mathematics, and Informatics, IPB University, Bogor, 16680, Indonesia)
Jaharuddin Jaharuddin (School of Data Science, Mathematics, and Informatics, IPB University, Bogor, 16680, Indonesia)

Zero : Jurnal Sains, Matematika, dan Terapan
Vol 10, No 2 (2026): Zero: Jurnal Sains Matematika dan Terapan

This study analyzes and optimizes user satisfaction in online marketplaces by integrating SERVQUAL, the Kano Model, Quality Function Deployment (QFD), and Mixed Integer Linear Programming (MILP). Data were collected from 167 Shopee users, with 107 valid responses analyzed. SERVQUAL measured service quality dimensions, the Kano Model derived satisfaction and dissatisfaction parameters, QFD translated user requirements into internal service measures, and MILP selected optimal improvement alternatives under an IDR 90,000,000 budget constraint. The results show that assurance had the highest SERVQUAL weight (0.207), followed by tangibility (0.201), reliability (0.200), empathy (0.198), and responsiveness (0.195). The Pearson correlation based HoQ approach selected customer service training, automatic delivery update features, and feedback completion incentives, with a total cost of IDR 85,000,000 and a deviation value of 0.191. These findings provide scenario based decision support to improve the quality of marketplace services.