Asia-Pacific Journal of Operational Research

We present a deterministic model that specifies lane direction in a multi-laned bridge that has a movable barrier that divides the two directions of traffic flow, in order to reduce congestion. A probabilistic dynamic programming formulation for a stochastic extension of the model is also presented. Analysis of the special structure of the dynamic programming formulation provides new insights into important aspects of certain traffic planning problems and represents a useful addition to the traffic network planner's toolkit. A case study involving the lane direction management of an actual bridge is also provided.

Redundancy in constraints and variables are usually studied in linear, integer and non-linear programming problems. However, main emphasis has so far been given only to linear programming problems. In this paper, an algorithm that identifies redundant objective functions in multi-objective stochastic fractional programming problems is provided. A solution procedure is also illustrated. This reduces the number of objective functions in cases where redundant objective functions exist.

In this paper, realistic production-inventory models with shortages for a deteriorating item with imprecise preparation time for production (hereafter called preparation time) has been formulated and an inventory policy is proposed for maximum profit in a finite time horizon. Here, the rate of production is constant, demand depends on selling price, marketing cost, and inventory level, and setup cost depends on preparation time. The imprecise preparation time is assumed to be represented by a fuzzy number and is first transformed to a corresponding interval number and then following interval mathematics, the objective function for total profit over the finite time horizon is changed to respective multi-objective functions. These functions are maximized and solved for a set of Pareto optimal solution with the help of the weighted sum method using the generalized reduced gradient technique. Different case studies have been made depending upon the occurrence of shortages. The models have been illustrated by numerical data. Pareto-optimal solutions for different weights are obtained and presented in tables and graphs.

In this article we analyze the M/G/1 retrial queue from a statistical viewpoint. Assuming that retrials are exponentially distributed we focus on the estimation of the retrial parameter for constant and non-constant retrial policies, including the non-ergodic case. We consider different estimators and compare their asymptotic variances. We test numerically the accuracy of the proposed estimators.

With the rapid advancement in computer technology, the software reliability plays a very significant role in predicting the performance and assuring the quality of the software. A compromise testing effort is necessary to affirm the pre-defined quality of the software at optimum cost by detecting and eliminating errors in the software. In this investigation, a robust solution for software reliability growth model (SRGM) which is applicable to non-perfect repair condition called generalized renewal process (GRP) is proposed. The GRP includes the renewal process (RP) and non-homogeneous Poisson process (NHPP) as special cases by suitable choice of index parameters. The optimal testing time of the software is determined for a large and complex modular software system wherein each module having a different failure rate and other characteristics. Maximum likelihood approach is used to estimate the GRP parameters. Numerical illustrations are provided to explore the effect of various parameters on maintenance cost.

The aim of this paper is to explore the efficiency and the benchmarks of financial holding companies (FHCs) for a small open economy, Taiwan. We employ a two-stage production process including profitability and marketability performance using a non-parametric frontier method — data envelopment analysis (DEA). Furthermore, the factor-specific measure and BCC (Banker–Charnes–Cooper) model are combined together not only to identify the inputs/outputs that are most important but also to distinguish those FHCs which can be treated as benchmarks. Our empirical result shows that (1) big-sized FHCs are generally more efficient than small-sized ones; (2) FHCs with the main body of insurance averagely perform better than the other two types (banks and securities); (3) while small efficient FHCs are easily to become benchmarks, big efficient FHCs are deemed as competitive niche players; (4) further mergers and acquisitions among FHCs should be considered so as to achieve economies of scale. The profitability/marketability matrix of FHCs is also presented.

This paper discusses a discrete-time Geo/G/1 retrial queue with the server subject to breakdowns and repairs. The customer just being served before server breakdown completes his remaining service when the server is fixed. The server lifetimes are assumed to be geometrical and the server repair times are arbitrarily distributed. We study the Markov chain underlying the considered queueing system and present its stability condition as well as some performance measures of the system in steady-state. Then, we derive a stochastic decomposition law and as an application we give bounds for the proximity between the steady-state distributions of our system and the corresponding system without retrials. Also, we introduce the concept of generalized service time and develop a recursive procedure to obtain the steady-state distributions of the orbit and system size. Finally, we prove the convergence to the continuous-time counterpart and show some numerical results.

Given an undirected graph with weights associated with its vertices, the minimum weight vertex cover problem seeks a subset of vertices with minimum sum of weights such that each edge of the graph has at least one endpoint belonging to the subset. In this paper, we propose a hybrid approach, combining a steady-state genetic algorithm and a greedy heuristic, for the minimum weight vertex cover problem. The genetic algorithm generates vertex cover, which is then reduced to minimal weight vertex cover by the heuristic. We have evaluated the performance of our algorithm on several test problems of varying sizes. Computational results show the effectiveness of our approach in solving the minimum weight vertex cover problem.