KIbaroP

AI-based robust production planning

The research project 'KIbaroP' aims to develop and validate an AI-based and robust approach to production planning with special consideration of SME-specific data.

Initial Situation

In manufacturing companies, the improvement of disruption avoidance plays a central role against the background of increasing susceptibility to disruptions in production. In industrial practice, disruptions are currently not taken into account in production planning, or only through estimated surcharges (z. e.g. 10 to 20 percent of the throughput time) on the planned throughput time. These surcharges are used in an attempt to increase adherence to deadlines despite any disruptions that may occur. However, there is a risk that excessively long lead times will lead to reduced customer satisfaction, higher storage costs and reduced productivity.

Expected Result

A robust approach to production planning based on artificial intelligence (AI) is being developed and validated, with special consideration of the specific challenges of small and medium-sized enterprises (SMEs), such as limited data availability. Disruptions from the areas of personnel, materials, equipment, warehouse, process and IT are taken into account in production planning on a situational and preventative basis. As a rule, existing data from production (e.g. movement data, machine data) is to be taken into account and aggregated with data from the operational system landscape in such a way that the information base on disruptions to the production system can be used in production planning. For this purpose, both historical and current feedback data (e.g. quality data, real-time data, current production sequence) are to be used to assess the risk of disruption to individual work processes. Based on this, time buffers for orders are dimensioned in such a way that possible disruptions can be compensated for in advance. The information obtained is processed using AI approaches in order to be able to make statements about critical orders or workstations with regard to their susceptibility to disruption, among other things. The results of the modeling are used for robust production planning.

Solution Approach

The solution approach is based on the industry-oriented process CRISP-DM (Cross Industry Standard Process for Data Mining) in order to develop a practical solution as a software demonstration application for SMEs. This application takes into account order data, feedback data and other data from the operational system landscape and the socio-technical environment. The data available for training AI algorithms forms the basis for the application of AI approaches. The approach therefore involves combining different data sources in order to obtain the most comprehensive database possible. First, literature-based correlations between feedback data from production and different types of faults are identified. These correlations are then validated in collaboration with the participating SMEs and used to build a generic simulation model that maps the relationships between disruption types and feedback data. The generic simulation model is then used to generate synthetic feedback data sets which, together with the feedback data from the SMEs, can be used as a training data set for the AI algorithms. Based on this data set, the next step is to develop an AI model for predicting failures and integrate it into the SMEs' existing PPS. The results will then be evaluated and used to develop a demonstrator. Overcoming challenges in production planning resulting from a small amount of data is achieved by combining the data provided by the SMEs with synthetic data, as this means that even a small database at the SMEs is sufficient.

Benefits for the Target Group

The development of AI-based and robust production planning offers companies both direct and indirect benefits. The direct benefit lies in the creation of an understanding of disruptions, which enables preventive measures to be taken in production planning. In addition, the existing knowledge of production data is further deepened and its preparation for AI applications is described. It also shows how insights can be generated from disruption data and used preventively.

The indirect benefit is the long-term increase in competitiveness through higher delivery reliability and customer satisfaction. The sustainable increase in productivity through reduced disruption enables companies to increase their profitability.

Topic Area

  • Production Management

Research Focus

  • Produktionsregelung

FIR Navigator

  • AI and Data Science
  • Production Planning and Control
  • JRF Guiding Topic

    • Society & Digitization
    • Industry & Environment

    Projectinfos

    Duration
    01.10.202330.09.2025
    Funding no.
    23054 N
    Funding information

    The Cornet project 23054 N of the Research Association FIR e. V. at the RWTH Aachen University, Campus-Boulevard 55, 52074 Aachen, is funded via the AiF within the framework of the Cornet program for the promotion of international projects of pre-competitive joint research for the benefit of small and medium-sized enterprises by the Federal Ministry for Economic Affairs and Climate Action (BMWK) on the basis of a resolution of the German Bundestag.