Mohammad M. Amini

Performance characteristics of alternative genetic algorithmic approaches to the traveling salesman problem using path representation: An empirical study

Abstract The potential of Genetic Algorithmic (GA) approaches for solving order-based problems including the Traveling Salesman Problem (TSP) is recognized in a number of recent studies. By applying various GAs, these studies developed a set of unresolved GA design and configuration issues. The purpose of this study is to resolve the conflicting GA design and configuration issues by (1) concentrating on the classical TSP; and (2) developing, implementing, and testing a complete set of alternative GA configurations, 144 GAs are developed and evaluated by solvinh 5000 TSPs. A carefully designed statistical experimental plan accompanied by rigorous statistical analysis isolate the most promising configurations and identify their effect on solution time and quality. Although, the emphasis is on the TSP, the final results are applicable to other order-based problems that use sequence encoding.

A hybrid heuristic for the generalized assignment problem

Abstract The objective of this study is to develop a hybrid heuristic (HH) for solving the generalized assignment problem (GAP) and conduct a computational comparison with the leading alternative heuristic approaches. HH is designed around the two best known heuristics: Heuristic GAP (HGAP) and Variable-Depth-Search Heuristic (VDSH). Previous performance characteristic studies have shown that HGAP dominates VDSH in terms of solution CPU time, while VDSH obtains solutions of 13% to 300% better quality within a ‘reasonable’ time. The main idea in this paper is to hybridize the two approaches, such that the inherent values of both heuristics are realized.

Greedy solutions of selection and ordering problems

Abstract The greedy method is a well-known technique for approaching problems involving the selection and/or ordering of a set of elements from a given set so as to optimize a specific objective function. Theoretical frameworks dealing with the optimality of greedy methods, including greedoids and matroids, consider a greedy rule known as the best-in greedy algorithm. Considering selection and ordering problems, the primary goals of this study are: (1) To present a generalization of the best-in greedy algorithm. (2) To present a necessary and sufficient condition for the optimality of the generalized best-in greedy algorithm. (3) To investigate the potential of the theoretical results in the characterization of heuristic design for NP-hard problems. In this pursuit, we apply the two aforementioned methods to solve the single-machine scheduling problem with sequence dependent set-ups where the objective is to minimize the total tardiness. An in-depth computational study indicates that the generalized best-in greedy algorithm can significantly outperform the best-in algorithm. (4) In addition, we present an efficient dynamic programming approach for a special class of greedoid structures that, under a sufficient (and realistic) condition, yields optimal solutions.

A robust heuristic for the Generalized Assignment Problem

The Generalized Assignment Problem, in the class of NP-hard problems, occurs in a wide range of applications — vehicle packing, computers, and logistics, to name only a few. Previous research has been concentrated on optimization methodologies for the GAP. Because the Generalized Assignment Problem is NP-hard, optimization methods tend to require larger computation times for large-scale problems. This paper presents a new heuristic,Variable-Depth-Search Heuristic (VDSH). We show that on the sets of large test problems, the quality of the solution found by VDSH exceeds that of the leading heuristic by an average of over twenty percent, while maintaining acceptable solution times. On difficult problem instances, VDSH provides solutions having costs 140% less than those found by the leading heuristic. A duality gap analysis of VDSH demonstrates the robustness of our heuristics.

Heuristic sensitivity analysis in a combinatoric environment: An exposition and case study

Abstract Sensitivity orpost-optimality analysis investigates the effect of parametric changes on heuristic robustness and solution quality. This approach is relatively unexplored for combinatorial optimization problems, and yet is of considerable interest in analyzing performance characteristics of heuristic approaches. The purpose of this paper is to: (1) develop the semantics and rationale of parametric analysis within the combinatoric environment; (2) present as an example the design and implementation of sensitivity analysis procedures for a newly developed heuristic — theVariable-Depth-Search Heuristic (VDSH) — to solve the Generalized Assignment Problem (GAP). The concepts and methodology discussed in this paper may as well be applied to other heuristics, or in developing a heuristic sensitivity analysis procedure for a large-scale optimization method.

A Simulation Analysis of Ordering Policies under Inflationary Conditions

In today′s competitive global economy, every business can benefit from managing its operations in a cost‐efficient manner. Managing materials should be no exception and the effect of inflation on ordering policies for purchased materials and supplies should be carefully analysed. It is very likely that a business can place itself in an advantageous position by changing its ordering policies in response to unfavourable economic conditions in its future operating environment. Presents a total cost model and, through simulation analysis, shows that inflationary conditions do affect lot sizes. Also shows that there are instances where lot sizes stay stable before changing again in response to increase in inflationary rates.

Supercomputing in corporate America: a sample survey

Abstract An information technology forecasts that supercomputing will have a significant impact on the data processing operations of corporations and be a competitive necessity in the 1990s. This study attempts to measure the degree of awareness, ability to use, adoption, and intention of corporate America to apply this new technology to improve their productivity and profitability, as well as their competitive edge in the global economy of the 1990s. We present the results of a study of 201 corporations in eighteen industries concerning their current and expected future use of supercomputers in mainstream business applications. Specifically, our purposes were to determine existing commercial supercomputing and report on the uses, awareness of the business potential of supercomputing, ability to “work with” supercomputers, utilization of commercial supercomputing, and perceived current and future supercomputer applications by competing firms.