2023: Testing a novel AI based productivity analysis and production optimization method in a production company

The objective of the project was to test the production performance at Trimtex and identify solutions to increase operational efficiency by optimizing order fulfillment and production cycle times. This was achieved through the validating of production flows using simulation methods and AI-based solutions in decision-making processes. 

A key focus of the project was the evaluation of the interfacing of the DIMUSA MES system into the Trimtex production process to enable real-time data collection, analysis, and optimization. The optimization recommendations based on the collected data help improve production efficiency, reduce costs, and ensure high product quality. The DIMUSA MES system enables seamless data exchange with other company systems, providing a comprehensive solution for managing production and supply chains. In summary, the data collected offers a detailed overview of the production process and allows for informed decision-making to improve efficiency and increase productivity. 

The validation of an AI-based optimization model and the application of cluster analysis based on production process data helps improve both the efficiency and quality of the production process at Trimtex. 

The analyses conducted in the project, the methodologies applied, and the use of interfacing capabilities between systems proved that this approach enables the analysis of production, identification of bottlenecks, and finding opportunities for improvement. Moving forward, the company can essentially choose different paths. By focusing on the DIMUSA MES system, it is possible to highlight cause-and-effect relationships and address production improvements and corrective actions retrospectively. 

By evaluating the interfacing AI technologies with the MES/ERP solution, the company could move towards a predictive approach in the entire order execution process framework, proactively generating order portfolios, production volumes, subunit load distributions, and equipment utilization. Additionally, this could allow for analyzing the efficiency of product mix throughput across the entire value chain.