Haluk Yapicioglu

Solving the semi-desirable facility location problem using bi-objective particle swarm

AbstractIn this paper, a new model for the semi-obnoxious facility location problem is introduced. The new model is com-posed of a weighted minisum function to represent the transportation costs and a distance-based piecewise function torepresent the obnoxious effects of the facility. A single-objective particle swarm optimizer (PSO) and a bi-objective PSOare devised to solve the problem. Results are compared on a suite of test problems and show that the bi-objective PSOproduces a diverse set of non-dominated solutions more efficiently than the single-objective PSO and is competitive withthe best results from the literature. Computational complexity analysis estimates only a linear increase in effort withproblem size. 2006 Elsevier B.V. All rights reserved. Keywords: Multiple criteria analysis; Semi-obnoxious; Location; Evolutionary computation; Bi-objective PSO 1. IntroductionThe semi-obnoxious facility location problem [5–8,17,30,37,41] has drawn much recent attention amongresearchers. Erkut and Neuman defined an obnoxious facility as one that generates a disservice to the peo-ple nearby while producing an intended product or service [17]. If minimization of the undesirable effects isthe only concern, transportation costs to/from the facility to be located may be prohibitive. Taking this intoaccount, Brimberg and Juel [8] introduced the term semi-desirable facility in 1998. They argued that thefacilities cannot be classified as being purely desirable or purely obnoxious. Garbage dump sites, airportsand power plants are typical semi-desirable facilities [8,17]. However, as noted in [17], the first paper that

Retail space design considering revenue and adjacencies using a racetrack aisle network

In this article, a model and solution approach for the design of the block layout of a single-story department store is presented. The approach consists of placing departments in a racetrack configuration within the store subject to area and shape constraints. The objective function considers the area allocated to each department, contiguity of the departments to the aisle network, adjacency requirements among departments, and department revenues. The revenue generated by a department is defined as a function of its area and its exposure to the aisle network. The aisle network is comprised of two components: the racetrack, which serves as the main travel path for the customers, and the entry/exit aisle. The racetrack aisle itself is treated as a department with area allocation and corresponding revenue generation. A general tabu search optimization framework for the model with variable department areas and an aisle network with non-zero area is devised and tested.

A conceptual architecture for static features in physical security simulation

The aim of this paper is twofold: first, to propose a data model that enables the user to model a physical facility at different levels of detail and explicitly incorporate interactions among the components of the facility. Second to suggest a methodology for line-of-sight, which is the primary factor in recognition of threats in physical security settings.

Bi-criteria model for locating a semi-desirable facility on a plane using particle swarm optimization

The problem of locating a semi-desirable facility on a plane is considered. A bi-criteria model is used. One of the criteria is well known minimum criterion. The second criterion is a weighted sum of Euclidean distances raised to the power of negative one. This function represents the aggregate undesirable effects of the facility and it is also a minimization problem. The bi-criteria model consists of a linear combination of these two criteria. The proposed model solved by particle swarm optimization provides better results than a previously proposed heuristic.

A bi-objective model for the retail spatial design problem

In this article, a model for the design of the block layout of a retail store is presented as a bi-objective optimization problem. The approach consists of placing departments in a racetrack configuration within the store subject to area and shape constraints, where the first objective function considers the area allocated to each department, contiguity of the departments to the aisle network and resulting revenue generated, and the second objective function considers adjacency requirements among departments. Owing to the complex nature of the model, two metaheuristic search methods, a multi-objective tabu search and the most well known multi-objective genetic algorithm are used separately to solve the problem. The performance of these two heuristics is evaluated and compared, with results suggesting that the multi-objective tabu search is a better choice because of its ability to exploit the neighbourhood structure of the model.

Hybrid approach for Pareto front expansion in heuristics

Heuristic search can be an effective multi-objective optimization tool; however, the required frequent function evaluations can exhaust computational sources. This paper explores using a hybrid approach with statistical interpolation methods to expand optimal solutions obtained by multiple criteria heuristic search. The goal is to significantly increase the number of Pareto optimal solutions while limiting computational effort. The interpolation approaches studied are kriging and general regression neural networks. This paper develops a hybrid methodology combining an interpolator with a heuristic, and examines performance on several non-linear bi-objective example problems. Computational experience shows this approach successfully expands and enriches the Pareto fronts of multi-objective optimization problems.

Neural Network Enhancement of Multiobjective Evolutionary Search

In this study, a novel approach is used to identify nondominated solutions to multiobjective optimization problems. The method is composed of a Particle Swarm Optimizer (PSO) coupled with a neural network. The PSO is used to find an initial set of nondominated solutions. These nondominated solutions are then used to construct a general regression neural network that generates a considerably larger set of nondominated solutions. Our neural network enhancement process is demonstrated on a test suite of six instances of bi-criteria semidesirable facility location problems. Results show that the set of nondominated solutions developed by the neural network is, on average, 25 times larger than the initial set found by PSO, and in many instances dominate those identified by PSO. The method developed within is straightforward and general and is a new alternative to multiobjective optimization with decision variables in continuous space.

A simulation methodology for online process control of hot mix asphalt (HMA) production

The quality of hot mix asphalt (HMA) is directly related to the quality of the input aggregates and the control of the production process. Many factors such as aggregate gradation and moisture level affect the quality of hot mix asphalt. As state agencies dictate certain standards on quality of the product, some quality assurance techniques have been used in HMA plants. In the current practice, a production sample is taken and analyzed in the lab. The lab analysis takes approximately two hours, making it difficult to quickly correct production mix problems. In this paper, a new online process control of asphalt production system designed to overcome this problem is described. In the proposed system, an image processing system continuously analyzes images of the samples and the required corrective action is taken instantly by a computerized optimization system. In this paper, the simulation model of the proposed online process control system is presented and the results are discussed.

Disservice representation using the Gini coefficient in semi-desirable facility location problems

We consider various bi-objective models for the semi-desirable facility location problem. In these problems, the disservice caused by the facility is traditionally measured by distance-related objective functions. In this paper, we modify the objective function representing the disservice using the Lorenz curve and the Gini coefficient. Both of these concepts are widely used in the economics literature to measure the discrepancy in wealth distribution within a population. The use of the Gini coefficient enables the measurement of how the disservice caused by the facility varies across different Pareto optimal solutions. We use a bi-objective particle swarm optimizer (bi-PSO) to compare how the change in the objective function representing the disservice affects the recommended location of the facility. Results suggest that some solutions identified as “Pareto optimal” by traditional formulations are dominated by other solutions when the Gini coefficient is used. Additionally, the use of the Gini coefficient causes a change in the “optimal” location of a semi desirable facility in some instances.