摘要：The complexity of process engineering projects in the chemical industry — resulting from the large number of activities to be accounted for as well as the required actors and resources — and the number of projects running simultaneously within an enterprise are rapidly increasing. In order to stay competitive, the factors relevant to the success of project planning and execution, e.g. the project budget or duration, must be accurately predicted and controlled. For this reason, a novel simulation approach for development projects is introduced and validated. A formal description of a development project and of an activity-oriented simulation model is given. This "meta model" is able to describe the influencing factors of a development project as well as their interrelations during the course of a project. On the basis of the meta model, an activity-oriented simulation model is developed in cooperation with enterprises from the chemical industry. The simulation model enables the automatic creation and prospective benchmarking of complex, detailed project plans. The dynamics of such a development project are represented as a stochastic Petri net, including Java functions. Organizational factors of a development project such as task scheduling, the limited availability of actors and tools or uncertainty regarding the effort required to solve a task can be systematically studied through simulation experiments. The results of these experiments assist project managers in understanding the influence of the quantity and characteristics of actors and resources on project performance. In the validation study, a chemical process design project in a large enterprise is considered and the external validity of the stochastic project model is analyzed.
Loading...
