An Integration of Rank Order Centroid, Modified Analytical Hierarchy Process and 0-1 Integer Programming in Solving A Facility Location Problem
Hadhramout province is the major producer of dates in The Republic of Yemen. Despite producing substantial quantity and quality of dates, the business losses are still high. The situation worsens with the widespread of the black market activities. Recently, the Yemeni government has issued an agreem...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2013
|
| Subjects: | |
| Online Access: | https://etd.uum.edu.my/3862/1/s90773.pdf https://etd.uum.edu.my/3862/7/s90773.pdf https://etd.uum.edu.my/3862/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Hadhramout province is the major producer of dates in The Republic of Yemen. Despite producing substantial quantity and quality of dates, the business losses are still high. The situation worsens with the widespread of the black market activities. Recently, the Yemeni government has issued an agreement stating the importance of building a date palm packaging factory as a resolution to the problems. Hence, this study aims to identify the best location for a date palm packaging factory among the seven districts which produce most of the date palm supplies in Hadhramout. The selection was based on eleven criteria identified by several representatives from the farmers and the local councils. These criteria were market growth, proximity to the markets, proximity to the raw materials, labor, labor climate, suppliers, community, transportation cost, environmental factors, production cost, and factory set up cost. The level of importance and the respective weight of each criterion were calculated using two different approaches, namely, Analytic Hierarchy Process (AHP) and Rank Order Centroid (ROC). In applying AHP, a slight modification was made in the pairwise comparison exercises that eliminated the inconsistency problem faced by the standard AHP pairwise comparison procedure. Likewise, in applying ROC, a normalization technique was proposed to tackle the problem of assigning weights to criteria having the same priority level, which was neither clarified nor available in the standard ROC. Both proposed techniques revealed that suppliers were the most important criterion, while community was regarded to be the least important criterion in deciding the final location for the date palm factory. Combining the criteria weights together with several hard and soft constraints that were required to be satisfied by the location, the final location was determined using three different mathematical models, namely, the ROC combined with 0-1 integer programming model, the AHP combined with 0-1 integer programming model, and the mean of ROC and AHP combined with 0-1 integer programming model. The three models produced the same result; Doean was the best location. The result of this study, if implemented, would hopefully help the Yemeni government in their effort to improve the production as well as the management of the date palm tree in Hadhramout. |
|---|