Planning and proper management practices in all plans and industrial projects are carried out in one of two forms; centralized or decentralized. Centralized projects, at all stages of planning, through execution and performance, rely on man power and the internal equipment of the company, whereas in decentralized projects, dependence on internal capabilities is reduced and efficiency is increased by the capabilities, equipment and contractors outside the company.In new systems of industrial management and projects, the trend towards decentralized projects has increased. Because of an increase in the extent and complexity of projects, and the inadequacy of inter organizational powers. The most important factor in the success of decentralized project management is the availability of a powerful information system for collecting and treating contractors' data and supporting the decision with a high level of confidence and desired accuracy. In this thesis, a supportive system of decision is presented that, including presentation of indexes of performance measurement and project progress and the designing of a decision tree required by the company, tries to assist the coordinanation, supervision and management of decentralized projects. To easily reach this goal, computer software, with a combination of capabilities, including a supporting informative system for decision making and knowledge of the management of decentralized projects, is designed and presented. One of the important duties of this software is supporting repeated and unrepeated decisions during the performance of a decentralized project by suggesting the optimal possible options with a separation of different levels of management, a way of treating time divisions, costs, the contractors,... etc.As a case study, MATN Oil Co. was selected, for providing the capabilities of this system, which is managed and planned as completely decentralized.