Thesis Archive
A Capacitated facility location model with stochastic demand for a fixed public server facility with the anticipation of siting an undesirable facility (2009)
Alexis D. Canete
Daniel Ford M. Magpantay
Abstract:
-Location models have been extensively studied since the 1960s, in the fields of operations research, management science, industrial engineering, economic geography and spatial planning literature. One of the questions that concern the decision makers is whether to increase the coverage of a service by expanding an existent facility or by building a new one. In building new facilities, facility location models are a great tool to justify the need for the new facility. Today, countless facility location models have been developed from the fundamental mathematical models.
It is evident that there is a lack of a mathematical model that considers integrating the factors of undesirable facilities with locating a public service facility. Most facility location models on public services consider modeling the desirable and undesirable facilities separately and does not account for the interaction between the two facilities. The given models fail to account the impact of the two different facilities with each other. In general, there is a need for a model that will take into account the effect of having these two types of facility in a given populated area with pre-existing or soon to raise undesirable facilities.
The purpose of this study is to help the decision makers in locating public service facilities while considering the preexisting and soon to rise undesirable facilities. The objective of the model is to minimize the total cost of capacity, traveling cost and penalty cost from undesirable facilities. The constraints include service level requirement, multi-facilities, single source, capacity, penalty cost, assignment, and non-negatively constraints. The model was validated through the use of GAMS, the inputs used were approximated from a published study. A small model was first created for the initial validation to check the feasibility of the model. If the small model would run and produce a feasible solution, then it will be expand into the intended actual model. If the small model is infeasible, the model has to be reviewed for errors and infeasibility. The model will be reformulated until there are no errors and infeasibility. The parameter values are also checked to see if the parameter values may have caused the infeasibility or illogical results. The results of the study revealed that there is a significant interaction between the two types of facilities, the public service facility and the undesirable facilities. The effect of having a minimum safe distance between the desirable and undesirable facility yielded to be significant. The parameters used were analyzed in terms of their sensitivity, and conclusions were derived from the sensitivity analysis.
For future studies it can be recommended to consider other factors that affect the patronage of a user, such as brand of facility, aesthetics, ambiance and service satisfaction. Another approach is to consider other elements in the community, such as other facilities that provides the same service. This would improve the current model since in locating desirable facilities with respect to the undesirable facilities, the effect of having a minimum safe distance between the desirable and undesirable facility yielded to be significant. Lastly, a single demand point can be served by multiple facilities.
Adviser:
Mr. Richard Li