For the task-scheduling problem, we propose an augmented neural-network approach, which allows the integration of greedy as well as nongreedy heuristics (AugNN-GNG), to give improved solutions in a small number of iterations. The problem we address is that of minimizing the makespan of n tasks on m identical machines (or processors), where tasks are nonpreemptive and follow a precedence order. The proposed approach exploits the observation that a nongreedy search heuristic often finds better solutions than do their greedy counterparts. We hypothesize that combinations of nongreedy and greedy heuristics when integrated with an augmented neural-network approach can lead to better solutions than can either one alone. We show the formulation of such integration and provide empirical results on over a thousand problems. This approach is found to be very robust in that the results were not very sensitive to the type of greedy heuristic chosen. The new approach is able to find solutions, on average, within 1.8% to 2.8% of the lower bound compared to 2.0% to 8.3% for the greedy-only AugNN approach. This improvement is obtained without any increase in computational complexity. In fact the number of iterations used to find the solution decreased.

This paper addresses the problem of selecting primary and secondary routes for every pair of communicating nodes in a backbone computer network. A primary route and a secondary route constitute a route pair. The secondary route is node-disjoint (and hence also link-disjoint) from the primary route. A mathematical programming model is developed whose objective is to select routes to minimize the mean delay faced by messages. The model captures situations where a single link or node failure would divert traffic to the appropriate secondary routes. A solution procedure based on a relaxation of the problem is presented. Computational results indicate the effectiveness of the solution procedure. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.

A conceptual framework for building a network of knowledge-based systems (KBS) to fully utilize the benefits of these systems within complex decision-making environments is proposed. The authors focus on three issues: (1) what should be the capabilities of the system supporting the human decision-maker; (2) how the nodes of such a network should be linked; and (3) what knowledge each node in the network should possess. A framework for modeling knowledge and communications between the nodes of a networked KBS is outlined. The framework is developed from the viewpoint that the decision support that is provided by the KBS is not restricted to the easy access to, and processing and management of, models and data. It also includes expertise, which the user does not possess and might need for the solution of the decision problems. The resulting conceptual framework draws on concepts from decision-support systems and expert systems.>

A model is developed that assigns primary and secondary (backup) concentrator coverage to each terminal site. The objective is to minimize communications costs as well as costs for setting up and operating the concentrators subject to capacity constraints. A relaxation of the problem is studied, and an effective solution procedure that makes the use of this relaxation is developed. Experimental results over a wide range of problem structures show that this solution procedure is very effective. It is also found to be significantly faster than a state-of-the-art commercial integer programming package.>

Design of distributed computer systems is a complex task requiring solutions for several difficult problems. Location of computing resources and databases in a wide-area network is one of these problems which has not yet been solved satisfactorily. Solution of this problem involves determining number and size of computer facilities and their locations, configuring databases and allocating these databases among computer facilities. An integer programming formulation of the problem is presented. Heuristic and optimal solution procedures are developed and computational experience with these procedures is reported. Implications of the model for designing distributed systems are discussed.