Tender Participants Selection Based on Artificial Neural Network Model for Alternatives Classification

Business-to-business, business-to-government, business-to-consumer tender support is considered. The problem of tender participants selection is stated as a separate stage of tendering process. It is the problem of tender alternatives classification on the tender offer demands accordance. The method of tender participants selection was proposed for the problem solving. The method is based on artificial neural networks application. It is proposed to use feedforward neural networks with a hidden layer. The results of expert evaluation of tender alternatives (project value, project due date, technical parameters etc.) are used as neural networks inputs. Neural network models configuration is realized based on evolutionary modeling heuristic approach which allows to use this method for tendering process support in different subject fields. Application of the proposed method allows to select the set of tender alternatives which should be financed together or should be used interchangeably if only one tender alternative is financed. Information technology in the form of web-based system was developed on the basis of client-server architecture. The experimental investigation of the proposed method was conducted for the tender participants selection problem solving in the building projects realization support. The received experimental results allow to recommend the proposed method for use in practice.