Jerzy Duda

Distributed evolutionary computing system based on web browsers with javascript

The paper presents a distributed computing system that is based on evolutionary algorithms and utilizing a web browser on a clients side. Evolutionary algorithm is coded in JavaScript language embedded in a web page sent to the client. The code is optimized with regards to the memory usage and communication efficiency between the server and the clients. The server side is also based on JavaScript language, as node.js server was applied. The proposed system has been tested on the basis of permutation flowshop scheduling problem, one of the most popular optimization benchmarks for heuristics studied in the literature. The results have shown, that the system scales quite smoothly, taking additional advantage of local search algorithm executed by some clients.

BUSINESS INTELLIGENCE AND NOSQL DATABASES

Since its introduction in the mid-1970s, the small computer has evolved into a powerful, easy-to-use device that is often a cost-effective solution to problems long troubling business professionals. MIS managers frequently are called on to establish procedures for or perform the actual acquisition process.

Advances in Fuzzy Decision Making

This book shows how common operation management methods and algorithms can be extended to deal with vague or imprecise information in decision-making problems. It describes how to combine decision trees, clustering, multi-attribute decision-making algorithms and Monte Carlo Simulation with the mathematical description of imprecise or vague information, and how to visualize such information. Moreover, it discusses a broad spectrum of real-life management problems including forecasting the apparent consumption of steel products, planning and scheduling of production processes, project portfolio selection and economic-risk estimation. It is a concise, yet comprehensive, reference source for researchers in decision-making and decision-makers in business organizations alike.

Lot-sizing in a foundry using genetic algorithm and repair functions

The paper presents a study of genetic algorithms applied to a lot-sizing problem, which has been formulated for an operational production planning in a foundry. Three variants of genetic algorithm are considered, each of them using special crossover and mutation operators as well as repair functions. The real size test problems, based on the data taken from the production control system, are presented for assessment of the proposed algorithms. The obtained results show that the genetic algorithm with two repair functions can generate good suboptimal solutions in the time, which can be acceptable from the decision maker point of view.

Collective intelligence of genetic programming for macroeconomic forecasting

A collective approach to the problem of developing forecasts for macroeconomic indicators is presented in the paper. The main advantage of genetic programming over artificial neural networks is that it generates human readable mathematical expressions that can be interpreted by a decision-maker. Gene expression programming used in the paper is an example of collective adaptive system, but we propose to use a collective intelligence to develop not only one forecasting model, but a set of models, from which the most suitable one can be chosen automatically or manually by the decision-maker.

Multiple criteria lot-sizing in a foundry using evolutionary algorithms

The paper describes the application of multiobjective evolutionary algorithms in multicriteria optimization of operational production plans in a foundry, which produces iron castings and uses hand molding machines. A mathematical model that maximizes utilization of the bottleneck machines and minimizes backlogged production is presented. The model includes all the constraints resulting from the limited capacities of furnaces and machine lines, limited resources, customers requirements and the requirements of the manufacturing process itself. Test problems based on real production data were used for evaluation of the different evolutionary algorithm variants. Finally, the plans were calculated for a nine week rolling planning horizon and compared to real historical data.

Differential evolution applied to large scale parametric interval linear systems

Differential evolution (DE) is regarded to be a very effective optimisation method for continuous problems in terms of both good optimal solution approximation and short computation time. The authors applied DE method to the problem of solving large scale interval linear systems. Different variants of DE were compared and different strategies were used to ensure that candidate solutions generated in the process of recombination mechanism were always feasible. For the large scale problems the method occurred to be very sensitive to the constraint handling strategy used, so finding an appropriate strategy was very important to achieve good solutions in a reasonable time. Real world large optimisation problems coming from structural engineering were used as the test problems. Additionally DE performance was compared with evolutionary optimisation method presented in [10].

A comparison of metaheurisitics for the problem of solving parametric interval linear systems

The problem of computing a hull solution of parametric interval linear systems with general dependencies is considered. It can be reduced to the problem of solving a family of constrained optimizatiom problems. In this study, different metaheuristics are used to solve those problems. Comparison of evolutionary algorithm, simulated annealing and tabu search algorithm together with analysis of variance tests are provided on the basis of three different practical problems.

A genetic algorithm for lot sizing optimization with a capacity loading criterion

The lot sizing problems are one of the basic optimization problems, which have to be solved during production planning. Only a few nature-inspired algorithms have been proposed for solving such problems. In this study the authors propose a genetic algorithm for a discrete lot sizing problem with so called small buckets and the criterion of capacity utilization. The results are compared with CPLEX MIP solver and other heuristics. The genetic algorithm proposed here gives solutions, which are 0.4 to 2.9 percent away from the theoretical lower bound.

A hybrid genetic algorithm and variable neighborhood search for multi-family capacitated lot-sizing problem

Abstract The paper presents a genetic algorithm (GA) hybridized with variable neighborhood search (VNS) to solve multi-item capacitated lot-sizing multi-family problem with setup times. The problem has a practical application in production planning e.g., in foundry industry, so test cases for computational experiments were based on the data from the real production process in a foundry. The VNS algorithm is used after a certain number of GA generations for all individuals in the population to improve solutions. The presented method applied for large instances of the problem outperforms both a dedicated genetic algorithm and a CLPEX Solver-based rolling horizon methods known from the literature.