SimCast

Short description - Custom plant engineering—a sector heavily dominated by small and medium-sized enterprises (SMEs)—can only to a limited extent derive process times from previous projects and apply them directly to new projects, given the customer-order- and component-specific nature of these projects. Consequently, due to uncertainties and potential disruptions, additional time buffers are often factored into scheduling; these are typically cost-intensive and can represent a competitive disadvantage for SMEs. The project management tools currently in use do not yet include a methodology to assist planners in reliably forecasting logistical processes in projects involving one-off and small-batch production. The goal of this research project is therefore to develop a methodology—based on expert knowledge and the analysis of historical project data—for creating a forecasting model to determine the duration of logistical processes. This forecasting model is intended to serve as a decision-support tool to achieve higher planning quality and thereby minimize risks in scheduling. The project is expected to yield a corresponding methodology and a process model; these will be implemented in the form of a demonstrator and validated through industrial applications and simulation. The methodology should be designed as a user-friendly add-on for project management tools already in use, so that it can be effectively implemented for operational use after the project ends.

Project Duration:
Mar 1, 2017 bis Nov 30, 2018

Project Management:
Prof. Dr. Christoph Laroque

Project Description

Tasks:

Based on the current state of research, the project’s fundamental working hypothesis is that existing historical data and the expert knowledge available within a company can be systematically formalized using an appropriate methodology and subsequently used to forecast the duration of the logistics process under consideration as part of project planning.

By externalizing the planning knowledge of the individuals involved in project planning processes—as knowledge carriers—and combining this with the systematic use of existing historical project data, the aim is to create a sound and sustainable basis for decision-making regarding future processes. The basic idea here is that the logistical processes involved in one-off and small-batch production can be classified, described in general terms using parameters to be specified within the scope of the project, and placed in a clear, quantifiable relationship to process duration. The values of these parameters can then be used to forecast process durations, thereby forming the basis for simulation-based validation of the overall project plan.

The primary objective of the methodology to be implemented is to provide a feature for forecasting process duration as a decision-support tool for the planners involved, complementing the project management tools currently used by SMEs. The methodology could later also be implemented as an extension of existing project management tools. The innovation lies in the practical application of existing data analysis methods, combined with externalized expert knowledge, thereby enabling improved decision support for SMEs and allowing the planning process as a whole to be carried out to a higher standard.

Project Objectives:

The goal of the research project is to develop a methodology for dynamically verifying process durations for reference logistics processes. The key innovation of the project lies in the practical and sustainable utilization of the company’s available project knowledge to support operational decision-making, particularly in the early planning phases of SMEs.

Specifically, the following objectives are to be achieved within the scope of the research project:

  • Forecasting model for logistics process durations for project planning by deriving and quantifying process parameter values from past projects
  • Procedural model for identifying significant influencing factors, such as the specific characteristics of the plant components of the respective one-of-a-kind unit, the delivery date, and the delivery location
  • Use of collective expert knowledge and statistical methods for data analysis (e.g., regression and correlation analyses, response surface models)
  • Ensuring the validity of the determined process durations through dynamic validation via simulation
  • Methodological expansion of the project management tools currently in use at SMEs
  • Evaluation of the developed methodology based on a demonstrator

IGF Project 19371 of the German Logistics Association (BVL) was funded by the Federal Ministry of Economics and Technology (BMWi) through the Alliance for Industrial Research (AiF) as part of the Program for the Promotion of Joint Industrial Research (IGF), pursuant to a resolution of the German Bundestag.