Smarter ERP-system

Important results from the project

The project has had two related objectives. The first goal has been to upgrade the ERP system MONITOR towards increased levels of predictive analysis and related systems support related to system users´ decision-making with the aim to increase the precision and quality of the planning in manufacturing firms. A second more overall goal has been to launch the work of developing the company Monitor ERP System AB´s expertise within AI/machine learning as to support future continued development efforts.

Expected long term effects

The project has focused on building a system capability in MONITOR with the support of AI to predict delays in production orders. Training data has consisted of historical orders incl. information about delays. A number of classification algorithms for supervised learning have been tested and evaluated on the problem: naive Bayes, decision trees, PSL-DA, PCA-LDA and neural networks. The results show that a delay prediction is possible to achieve and that up to 40% of the delays in the test data can be predicted using the evaluated classification algorithms.

Approach and implementation

The project ran during the period August 2020 to January 2021. The work was carried out according to agile methodology and was conducted in teams with staff from RISE and Monitor. In this way, the project has generated concrete learnings about the potential of AI related to increased insight functionality of the system, and to staff at Monitor having increased their competence in AI/machine learning. General competence development has also taken place at three workshops where RISE presented various aspects of AI for key employees at Monitor.