طراحی و روش شناسی کاربردی برای برنامه ریزی تولید سلسله مراتبی سیستم های پشتیبانی تصمیم گیری در زمینه یکپارچه سازی سازمانی

Literature review on hierarchical production planning (HPP) shows that a methodology for the design and implementation of such systems is necessary. An overview of its main advantages and limitations is presented. At each stage of the methodology critical aspects are highlighted and some solutions are provided. The elements of the HPP decision support systems and its operation are described in an enterprise integration context and taking profit from the new information technologies.

Production planning [1] is a complicated task which requires cooperation among multiple functional units in an organization. For solving the problem of production planning powerful optimization models have been constructed by means of the formulation of mathematical programming. The result [2] may be a large scale model that is difficult to assemble, optimize or interpret as a single, monolithic entity. But the effort needed to record and update all data required by monolithic models, the difficulty in solving large mixed-integer programs as well as their assumption of a centralized decision-making, neglecting established organizational structures, obstructs the applicability of this type of models. A classical approach to overcome these difficulties of implementation is hierarchical production planning (HPP). The idea of HPP and scheduling was initiated by Hax and Meal [3]. The HPP approach recognizes and represents the planning process by a series of mathematical models and partitions the decision process into modules or subproblems with different planning time horizons. It also aggregates and disaggregates information through the various hierarchical levels. To ensure effective decision-making, a strong linkage must exist between these models at each hierarchical level. In this paper we analyze (Section 2) the advantages and limitations of existing HPP systems. In Section 3 some concepts of enterprise integration are outlined. In Section 4 we establish a design and application methodology for HPP decision support systems in an integration enterprise context. This methodology tries to avoid some critical design and implementation aspects of such a system, analyzing the different approaches in the existing literature and proposing alternative methods. The purpose of this paper is to present the methodology for theoretical foundations. An application to the tile industry is nowadays being developed and the results will be provided in the sequel.

A general framework of the actual HPP state has been presented, highlighting its strengths and weaknesses. To avoid the situation dependent shortcoming a design and application methodology for hierarchical production planning systems in an enterprise integration context (GRAI) has been presented. The difference between the designer and user has been established. It has been shown that the HPP module must be designed in a way that the production planner (end user) does not require any mathematical/modelling background to use the system. A description of the elements constituting the HPPDSS and their necessary vertical and horizontal integration has been made. Relating to vertical interdependencies, a not pure top-down one is recommended to avoid infeasibilities occurring at lower levels. Instead, some form of cooperation between levels is enhanced (anticipation of the lower level and iterative aggregation/disaggregation schemes). An updating weight procedure and the introduction of the slack have been proposed. However, to provide HPPDSS with more flexibility it is necessary to consider horizontal integration among the elements (a little investigated issue). In case unexpected internal or external events occur the decision maker can modify the plan with the least disruption. Although an HPPDSS provides the decision maker with a powerful mathematical base, a full control of the system by the decision maker must exist. Furthermore, such interactive computerized planning system (HPPDSS) is adaptable (flexible) to the changing environment through the integration with the information system. The way in which new information technologies can be used to support the decision-making process has been described.