MODELING LOGISTIC ENTERPRISE RE-LOCATION DECISION BY A NESTED LOGIT MODEL

Y Nguyen CAO

Dr.Eng Vietnam-Japan Research and Development Center, Department of Transport Economics,University of Transport and Communications,Cau Giay, No.03 Lang Thuong, Dong Da, Ha Noi, Viet Nam

ynguyencao82@utc.edu.vn

Abstract

This paper develops a model to analyze decisions regarding the relocation process for logistics enterprise by using nested logit model. In this framework, two decision points in the relocation process are assumed and maintained in the micro-simulation modeling. The first decision, move or non move and the second decision, choosing the destination location. This study applied the relocation decision structure of each logistics enterprise by nested logit model to find out the best model. In case study, the logistics enterprise relocation decision model has acceptable performance by the nested logit model. However, the nested logit model has to follow the IID Gumbel distribution holds within each nest. Therefore, nested logit model cannot take into account the various tastes among alternatives in the random part of utility function to improve the implementation of the model. The results indicate that big logistics enterprises have a lower probability of relocating and the migrating enterprises are more attractive in the zone which has a high accessibility. Finally, the population density, number of employees and the average land prices of zone strongly affect on the relocation decision making process of individual logistics enterprises.

Keywords: Logistics Enterprise, Re-location Decision Model, Nested Logit Model

read more

MODELING LOGISTIC ENTERPRISE RE-LOCATION DECISION BY A NESTED LOGIT MODEL

Y Nguyen CAO

Dr.Eng Vietnam-Japan Research and Development Center, Department of Transport Economics,University of Transport and Communications,Cau Giay, No.03 Lang Thuong, Dong Da, Ha Noi, Viet Nam

ynguyencao82@utc.edu.vn

Abstract

This paper develops a model to analyze decisions regarding the relocation process for logistics enterprise by using discrete choice models. In this framework, two decision points in the relocation process are assumed and maintained in the micro-simulation modeling. The first decision, move or non move, is modeled by using a binary logit form with outcome the probability of moving. The second decision, choosing the destination location, is modeled by a mixed logit model incorporating spatial effects with the outcome of the conditional probability of choosing a zone. This study also applied the relocation decision structure of each logistics enterprise by nested logit model to find out the best model. In case study, the logistics enterprise relocation decision model has acceptable performance by the nested logit model. However, the nested logit model has to follow the IID Gumbel distribution holds within each nest. Therefore, nested logit model cannot take into account the various tastes among alternatives in the random part of utility function to improve the implementation of the model. The proposed model also confirm again the important role of spatial interactions among individual logistics enterprise and among zones in the logistics enterprise relocation decision process. The results indicate that big logistics enterprises have a lower probability of relocating and the migrating enterprises are more attractive in the zone which has a high accessibility. Finally, the population density, number of employees and the average land prices of zone strongly affect on the relocation decision making process of individual logistics enterprises.

Keywords: Mixed Logit Model, Logistics Firm, Re-location Decision Model, Nested Logit Model

JEL classification:

read more