A model for warehouse performance measurement

Abstract

The traditional models for measuring warehouse performance have been used by single metrics and single productivity ratio such as fill rate per hour. While these traditional models are very easy to implement, they have been lacking in benchmarking performance over time. Therefore, this study aims to overcome such limitations of the traditional models by exploring the unknown relationships between performance indicators in five measurement areas: labour, cost, time, utilisation, and administration, throughout five typical steps in warehouse functions (receiving, handling, storing, order picking, and shipping). This study consists of two parts: quantitative study and qualitative study. In quantitative study, the proposed model was generated using the structural equation modeling (SEM) technique together with a dataset of 201 warehouses from hard disk drive, animal feed, and foam packaging industries in Thailand. With the use of the AMOS software program, the results show that the SEM model consists of ten indicators that are significant to warehouse performance at level of significance of 0.05. These indicators are, for example, labour hour, transportation cost, degree of space utilisation, and degree of error-free orders shipped. Importantly, the SEM model was tested whether it was acceptable based on the goodness-of-fit measures. The results show that the values obtained for the five measures: chi-square (p = 0.217), GFI = 0.972, NFI = 0.971, CFI = 0.995, and RMSEA = 0.032 indicating that the SEM model was acceptable to measure warehouse performance. Furthermore, the SEM model was extended using an analogy based approach. This approach allows a company to retrieve other warehouses’ information in order to improve their own performance. In qualitative study, the methodological triangulation was used to triangulate three different methods for warehouse performance measurement. These three methods are the SEM model, the multiple case study, and validation of the SEM model using data from 80 companies in Thailand and a company in Australia. The results show that all ten indicators from the SEM model are used by the Thai and Australian companies to measure their warehouse performance. Therefore, it can be concluded that the SEM model is able to measure warehouse performance in real situations.