Scheduling of Construction Projects with a Hybrid Evolutionary Algorithm’s Application

Abstract

In the works [5,16] there is presented a possibility of application of critical chain scheduling /buffer management methodology (CCS/BM) in investment enterprise and construction planning. The completion of a civil structure or a building complex is an undertaking consisting of the following factors: fulfilling the requirements (quality), cost, time of execution, range and resources [7]. In the article there are presented the outcomes of the tests concerning an improvement of methods used for investment scheduling and construction project with the implementation of mementic algorithm (i.e. hybrid evolutionary, HEA, [1]). The aim of the work was finding an optimal (for a taken goal function) level of workers’ employment that is minimization of a divergence from the average level of employment with the implementation of CCS/BM methodology [5]. Scheduling of investment and construction project is connected to an optimization task. It is related to finding the best solution fulfilling the constraining conditions and taking into consideration the goal function. There are many known methods of optimization applied in specific cases. Among others, there can be mentioned, for instance, for continuous tasks – methods of linear simplex, tasks of global optimization – when a goal function in the field of accepted solutions has more than one local minimum, discrete tasks with a greater complexity of calculations based most commonly on division or constraint methods, non – determined methods using random generating of solutions, a simulated annealing method as a modification of random walk with the improvement of quality of goal function, tabu search, that is with a list of revised variants and others. There are also techniques applied with the use of biological systems – evolutionary algorithms, genetic, evolutionary strategies, evolutionary programming and genetic programming ([3],[18]). The general scheme of evolutionary algorithm’s operation resides in creating a loop embracing reproduction, genetic operations, evaluation and succession. The classic scheme of operation of evolutionary algorithm is presented below according to [3]. This paper is a continuation of the topic presented in work [15], concerning the application of genetic algorithms [7] to steering of a level of an employment in investment and construction projects. Treating the evolutionary algorithm as a typical method of proceeding concerning