A Comparison Between Radial and Non-Radial Data Envelopment Analysis (DEA) Models: A Case of Trawl Vessels in Malaysia Coastal Seas

Abstract

In Malaysia, the consumption of fish-based protein is becoming increasingly popular in line with the growing population. However, the mismatch between fish production and the high demand for fish has led to the categorization of the fishing industry in Malaysia as inefficient and in a critical state. Among the factors contributing to this inefficiency are overfishing and the use of destructive vessels such as trawlers. In this study, the objective is to find the efficiency of trawlers for fishing was assessed in 12 coastal areas in Malaysia using two basic approaches with very different characteristics known as radial and non-radial Data Envelopment Analysis (DEA) models. The radial model is represented by the Charnes, Cooper, and Rhodes (CCR) model and the Banker, Charnes, and Cooper (BCC) model while the non-radial model is denoted by the Slack-Based Measure (SBM) model. Furthermore, a comparison of efficiency scores among the models was conducted to assess consistency between them. The Spearman’s Rank Correlation test revealed that the CCR and SBM models were significantly highly correlated. This shows that there is no significant difference between the radial model and the non-radial model for this case. The study found that the most efficient Decision-Making Unit (DMU) was Pulau Pinang, while the least efficient DMU was Johor Timur.