To content
Faculty of mechanical engineering

Data Science in Industrial Engineering

03/01/2023 - 02/28/2026

AI assistance system for plausibility checks, correction and creation of production plans

Problem

Flexible planning and adaptability are crucial for companies to succeed in today's dynamic global market landscape. High planning quality is a decisive economic factor, especially in the context of manual jobs, which continue to play an important role. AI-supported assistance systems offer a promising solution by extracting expert knowledge and suggesting solutions to complex problems

Graphic overview of a process chain from product definition through series development to series support. Multiple steps, such as product structure, process planning, and process description, are shown. An AI assistance system provides suggestions and performs plausibility checks. © IPS​/​TU Dortmund

Objective

The aim of the project is to support planners in the efficient creation of new production plans using an AI-based assistance system. This system will perform plausibility checks, automatically correct errors and support the creation of production plans with suggestions. This increases the quality of the plans and reduces the time required to create them. At the same time, the spread of incorrect data is minimized. Experts receive targeted support by being provided with relevant information and their specialist knowledge is optimally integrated. As a result, both existing and newly created production plans are designed precisely and efficiently

Schematic representation of an AI-based system for creating manufacturing plans. Several process steps, such as plausibility checking, correction, and training data flow, are connected by arrows. Expert knowledge and manual feedback complement the process between design and production. © IPS​/​TU Dortmund

Approach

The project uses a combination of rule-based evaluation and machine learning methods to determine plausibility assessments and corrections for existing production plans. Plans with low plausibility are automatically improved, while highly plausible plans serve as the basis for generating new plan proposals. The overall system consists of three core components that are being developed as part of the project:

Plausibility check to evaluate existing production plans
Correction to improve faulty production plans
Proposal system for generating new production plans

Related Links