Showing papers on "Heuristic (computer science) published in 1987"

Scheduling Tasks with Resource Requirements in Hard Real-Time Systems

Abstract: This paper describes a heuristic approach for solving the problem of dynamically scheduling tasks in a real-time system where tasks have deadlines and general resource requirements. The crux of our approach lies in the heuristic function used to select the task to be scheduled next. The heuristic function is composed of three weighted factors. These factors explicitly consider information about real-time constraints of tasks and their utilization of resources. Simulation studies show that the weights for the various factors in the heuristic function have to be fine-tuned in order to obtain a degree of success in the range of 75-88 percent of that obtained via exhaustive search. However, modifying the approach to use limited backtracking improves the degree of success substantially to as high as 99.5 percent. This improvement is observed even when the initial set of weights are not tailored for a particular set of tasks. Simulation studies also show that in most cases the schedule determined by the heuristic algorithm is optimal or close to optimal.

On Sequencing Retrievals In An Automated Storage/Retrieval System

Abstract: This paper addresses throughput improvement by retrieval sequencing in conventional unit load automated storage/retrieval systems when several retrieval requests are available and dual command cycles are performed. Taking first-come-first-served as the reference sequencing rule, the potential for improvement is identified. A “nearest-neighbor” sequencing rule is proposed as an alternative, an analytic model for its expected performance is developed, and Monte Carlo simulation is used for evaluation. In addition, a lower bound on dual command cycle times is developed, and the dynamic behavior of two heuristic sequencing rules is discussed.

Simulated annealing methods with general acceptance probabilities

Abstract: Heuristic solution methods for combinatorial optimization problems are often based on local neighborhood searches. These tend to get trapped in a local optimum and the final result is often heavily dependent on the starting solution. Simulated annealing methods attempt to avoid these problems by randomizing the procedure so as to allow for occasional changes that worsen the solution. In this paper we provide probabilistic analyses of different designs of these methods.

The node-weighted Steiner tree problem

Abstract: The general Node-Weighted Steiner Tree problem is an extension of the standard Steiner Tree problem by the addition of node-associated weights. This article analyzes a special case of that problem, where the set of nodes, which must be included in the solution tree, consists of a single node, and all node weights are negative. The special case is shown to be NP-Complete, its integer programming formulation is presented, and heuristic procedures are proposed. Using Lagrangian relaxation and subgradient optimization, tight lower bounds were derived and utilized by a branch and bound algorithm. The effectiveness of the developed procedures is demonstrated by a set of computational experiments.

Exact and Approximate Solutions for the Gate Matrix Layout Problem

Abstract: We consider the gate matrix layout problem for VLSI circuits, which is known to be NP-complete. We present an efficient algorithm for determining whether two tracks suffice. For the general problem of minimizing the number of tracks (and, hence, the area) needed, we design an attractive dynamic programming formulation to guarantee optimality. We also investigate the performance of fast heuristic algorithms published in the literature and demonstrate that there exist families of problem instances for which the ratio of the number of tracks required by these heuristics to the optimal value is unbounded. Moreover, we show that this result holds for any on-line layout algorithm. We additionally prove that, unless P = NP, no polynomial-time layout algorithm can ensure that the number of tracks it requires never exceeds k plus the optimum, for any constant k.

A dual-based algorithm for multiproduct uncapacitated facility location

Abstract: The multiproduct uncapacitated facility location problem (MUFLP) is a generalization of the classic uncapacitated facility location problem (UFLP). In MUFLP, different products are required by the customers. In addition to the fixed cost for opening a facility, there is an added fixed cost for handling a particular product. Assignment costs are incurred for satisfying a customer's requirement for each of the separate products. The objective is to minimize the total fixed costs plus assignment costs subject to satisfying all customer requirements. We propose a new dual-based algorithm for MUFLP that extends previous work for UFLP that has proved so successful. Dual ascent and dual adjustment procedures generate a good feasible solution to the dual of the linear programming relaxation of MUFLP. A feasible primal solution to MUFLP can then be constructed based on this dual solution. Like the analogous procedures for UFLP, these procedures can be used either as a stand-alone heuristic or, else, they can be in...

Time Slot Assignment in SS/TDMA Systems with Intersatellite Links

Abstract: In this paper we study the time slot assignment problem in clusters of SS/TDMA satellite systems interconnected through intersatellite links. We show that the problem of finding an assignment which minimizes the total transmission time is NP-complete, i.e., computationally intractable, even for quite restricted intersatellite link patterns and simplified system models. Successively, we focus our attention on clusters of two satellites, proposing a branch-and-bound optimal algorithm and two fast heuristic algorithms. We investigate the performance of the proposed heuristic algorithms both by a theoretical worst case bound and by simulation trials showing that the produced solutions are close to the optimal on the average.

The Product Matrix Traveling Salesman Problem: An Application and Solution Heuristic

Abstract: In this paper we analyze an optimization problem that arises in the overhaul of a gas turbine engine. This problem involves the placement of nozzle guide vanes in the nozzle of the engine. The objective of the vane placement is to attain "uniform" flow equalized distribution of hot fuel gases about the circumference of the engine nozzle. We show that this placement problem is equivalent to a traveling salesman problem TSP whose cost matrix is a product matrix. Exploiting properties of the special form of the cost matrix, we give a heuristic algorithm for solving the problem, and derive a posteriori lower bounds for the heuristic. We show that the heuristic performs extremely well on both real and simulated data. Finally, we present and develop the theoretical results that are the foundation of the proposed heuristic.

A decentralized approach to the motion planning problem for multiple mobile robots

Abstract: In this paper the motion planning problem for multiple mobile robots is addressed. Conventional methods of planning the motion for a single moving object are based on the assumption of a static environment, and so they cannot be used here because each of the robots is in a dynamic environment consisting of other moving robots. Centralized approaches to the multiple moving objects problem were shown to be intractable. In order to find a practical solution for the problem, it is necessary to reduce the complexity of it by decomposing the problem and introducing various heuristic techniques. We are proposing here a decentralized approach which is based on the decomposition of the problem into two subproblems: the global path planning problem and the local path replanning problem. This approach is based on a framework of problem solving using a group of intelligent agents.

Case-based problem solving with a large knowledge base of learned cases

Abstract: Recent experiments indicate that a case-based approach to the problem of word pronunciation is effective as the basis for a system that learns to pronounce English words. More generally, the approach taken here illustrates how a case-based reasoner can access a large knowledge base containing hundreds of potentially relevant cases and consolidate these multiple knowledge sources using numerical relaxation over a structured network. In response to a test item, a search space is first generated and structured as a lateral inhibition network. Then a spreading activation algorithm is applied to this search space using activation levels derived from the case base. In this paper we describe the general design of our model and report preliminary test results based on a training vocabulary of 750 words. Our approach combines traditional heuristic methods for memory organization with connectionist-inspired techniques for network manipulation in an effort to exploit the best of both information-processing methodologies.

A Parallel Heuristic Algorithm for Lot-Sizing in Multistage Production Systems

Abstract: Existing studies on heuristic lot-sizing techniques for multistage production systems concentrate on optimizing the production operations of the individual stages. This approach induces a family of heuristics, in which the production plans for the individual stages are computed sequentially starting from the final stage. In this paper we introduce a parallel heuristic for the problem. The method, which may be considered as the generalization of the single stage Wagner-Whitin algorithm, attempts to optimize over all stages simultaneously, while building the production plans in a forward manner. In computational experiments the new algorithm performed extremely well.

A Dual-Cost Heuristic For The Capacitated Lot Sizing Problem

Abstract: This paper examines a mathematical programming method of accounting for capacity costs for the deterministic, multi-item, single operation lot sizing problem. With the capacity constraints of CLSP removed with Lagrangian relaxation, the problem decomposes into a set of uncapacitated single product lot sizing problems which are solved with dynamic programming. The Lagrangian dual costs are updated by subgradient optimization. Feasible solutions (production plans within the capacity limitations) are constructed with a heuristic smoothing procedure. The dual-cost heuristic gave solutions which were better on average than the other algorithms tested (and was faster than some comparable algorithms).

Partitioning concurrent VLSI simulation programs onto a multiprocessor by simulated annealing

Abstract: Efficiently loading concurrent programs onto multiprocessor architectures is a graph partitioning problem where both the edges and vertices are weighted. The corresponding optimisation problem is computationally NP-hard, and the optimal solution can only be found by exhaustively examining all possible partitioning configurations. Near-optimal solutions can be found by using heuristic algorithms such as iterative improvement and simulated annealing. The simulated annealing heuristic is experimentally evaluated against simple iterative improvement for graphs representing the concurrent simulation programs of four VLSI circuits where the vertices were weighted. A simple cost function and an annealing schedule are presented for partitioning the graphs onto a star network of identical processors. Experimental results show that simulated annealing produces a better solution than simple iterative improvement but at the expense of considerable computer running times. It is suggested that the time required for simulated annealing to give better solutions than iterative improvement depends on the nature of distributions of weighted vertices and edges in the concurrent simulation graph as well as its size.

A heuristic algorithm for the periodic scheduling and sequencing job-shop problem

Abstract: This paper is devoted to the analysis of the job-shop scheduling and sequencing problem for a repetitive production process. An heuristic algorithm is described to find a near optimal solution such that the bottleneck machines are fully utilized in steady-state with a minimal number of jobs in-process. The developments are carried out using a modelling based on Timed Petri Nets.

A nonlinear programming heuristic for computing optimal link capacities in a multi-hour alternate routing communications network

Abstract: We consider the problem of determining the optimal number of communications channels in a telecommunications network. Traffic between pairs of nodes is assumed to vary with the time of day. Traffic between two nodes can flow over a one-link direct path or, if no direct channels are available, over a two-link alternate route. The problem is to compute the optimal number of channels on each link to minimize cost, subject to a constraint that enough trunks are present so that the probability of no idle channel does not exceed a specified amount. The mathematical formulation of this problem, known as the multi-hour engineering problem, yields an integer nonlinear programming problem NLP. We apply recursive quadratic programming, with exact penalty function EPF line searches, to solve the continuous NLP, and use the special problem structure to provide a starting point and EPF parameter. A simple heuristic is used to obtain integral number of channels satisfying desired blocking objectives. The method successfully solved a set of randomly generated test problems with up to 35 variables and 432 nonlinear constraints.

Validation of the first step of the heuristic refinement method for the derivation of solution structures of proteins from NMR data.

Abstract: A new method for the analysis of NMR data in terms of the solution structure of proteins has been developed. The method consists of two steps: first a systematic search of the conformational space to define the region allowed by the initial set of experimental constraints, and second, the narrowing of this region by the introduction of additional constraints and optional refinement procedures. The search of the conformational space is guided by heuristics to make it computationally feasible. The method is therefore called the heuristic refinement method and is coded in an expert system called PROTEAN. The paper describes the validation of the first step of the method using an artificial NMR data set generated from the known crystal structure of sperm whale carbon monoxymyoglobin. It is shown that the initial search procedure yields a low-resolution structure of the myoglobin molecule, accurately reproducing its main topological features, and that the precision of the structure depends on the quality of the initial data set.

Applying AI Techniques to Program Optimization for Parallel Computers

Abstract: TIIis paper describes an experiment of integrating expert systems technology and advanced compiler optimization teclmiques to the problem of paralIelizing programs for different classes of parallel computers. Our approach to solve the problem is to separate machine features from programming heuristics and organize the program parallelization knowledge in a hierarchical structure which we called heuristic hierarchy. The reasoning mechanism of the program restructuring system utilizes the heuristic hierarchy and features of the program and the target machine to choose appropriate sequences of transformations automatically. Theories and mechanisms for organizing and integrating the parallelism optimization knowledge are discussed. Methodologies for abstracting machine features, data management, and programming parallel computers are presented.

Accelerated branch exchange heuristics for symmetric traveling salesman problems

Abstract: A method for accelerating the computational performance of branch exchange heuristics for symmetric traveling salesman problems (TSP's) is presented. The improvement in performance is obtained by considering only exchanges that have a good chance of producing a better solution. In the instance of the 3–optimal heuristic, the approach reduces the number of operations required to obtain a good solution to a TSP with N nodes from O(N3) to O(N2), without a corresponding degradation in the quality of the solution. Most of the results are produced for Euclidean TSP's, but evidence is presented that indicates that thes results apply equally well if not more strongly to the general symmetric TSP.

Heuristic Algorithms for Single Row Routing

Abstract: A heuristic algorithm, based on the criterion of having nets with larger cut numbers assigned to inner tracks and nets with smaller cut numbers assigned to outer tracks, for single row routing problem has recently been proposed by Tarng et al. It has been reported that this algorithm has always been able to produce the optimal solutions for all the examples tested so far. In this paper, we have proved that algorithms based on the heuristic criterion of cut numbers produce optimal solutions for instances in which all nets cover at least one common node (i.e., form a single group). However, the algorithm proposed by Tarng et al. may not produce optimal solutions for instances of multiple net groups. Thus, several possible heuristic algorithms based on the same criterion, but also taking into consideration the net grouping situation have been proposed. The experimental results show that the proposed algorithms are faster and often generate better results than the one proposed by Tarng et al. A tighter lower bound on the number of tracks required is also obtained in this paper.

Multi-Item Scheduling by Benders' Decomposition

Abstract: Scheduling the production of several items requires the determination of production quantities in different periods in the presence of resource constraints. Several approximate and heuristic algorithms have been proposed to solve this problem. However, no method for finding an optimal solution has as yet been developed. It is shown that the problem may be solved advantageously using Benders' decomposition. The subproblem in Benders' decomposition is shown to be a transportation problem, and some strategies for solving the master problem are indicated. The paper concludes with a sample problem demonstrating the application of the method.

A Heuristic Algorithm for Gate Assignment in One-Dimensional Array Approach

Abstract: In this paper, we present a new approach for the one-dimensional gate assignment problem. The original minimization problem is transformed into a restricted problem, and then a new heuristic algorithm is applied to it. The solution obtained by the algorithm is interpreted as a solution for the original problem. The whole process of the approach has been implemented and tested with various examples. Experimental results show that our approach can approximately produce optimum solutions.