Innovative and Cognitive Production Technology and Systems, 16 - 18 July 2025
Home2021 Assembly & Battery ProductionTowards an intelligent disruption management system based on the maximal network plan – Development of a prioritization algorithm for disruptions in production processes
Towards an intelligent disruption management system based on the maximal network plan – Development of a prioritization algorithm for disruptions in production processes
thank you for your very interesting presentation dealing with disruption management based on the maximal network plan technique.
In your presentation, the combination between machine learning and expert knowledge is mentioned. Could you provide some more description about the kind of machine learning techniques utilized and the way it is combined with expert knowledge?
so far, we used regression techniques and SVMs to estimate the duration of troubleshooting, a parameter in our proposed method, that we even left out in our final tests.
In our paper, we describe that further machine learning techniques will be necessary for opimisation of the prioritisation method.
We used and will use expert knowledge to determine the relevant parameters that influence the criticality of disruptions and their possible weights. To verify and concretise these rather vague estimations, we will use machine learning techniques.
thank you for your question. So far, we used regression techniques and SVMs to estimate the duration of troubleshooting, a parameter in our proposed method, that we even left out in our final tests.
In our paper, we describe that further machine learning techniques will be necessary for opimisation of the prioritisation method.
We used and will use expert knowledge to determine the relevant parameters that influence the criticality of disruptions and their possible weights. To verify and concretise these rather vague estimations, we will use more machine learning techniques.
If you have any further questions, do not hesitate to contact us.
Dear Mr. Wagner,
thank you for your very interesting presentation dealing with disruption management based on the maximal network plan technique.
In your presentation, the combination between machine learning and expert knowledge is mentioned. Could you provide some more description about the kind of machine learning techniques utilized and the way it is combined with expert knowledge?
Thanks and kind regards.
Roberto Teti
Dear Mr. Teti,
so far, we used regression techniques and SVMs to estimate the duration of troubleshooting, a parameter in our proposed method, that we even left out in our final tests.
In our paper, we describe that further machine learning techniques will be necessary for opimisation of the prioritisation method.
We used and will use expert knowledge to determine the relevant parameters that influence the criticality of disruptions and their possible weights. To verify and concretise these rather vague estimations, we will use machine learning techniques.
Dear Mr. Teti,
thank you for your question. So far, we used regression techniques and SVMs to estimate the duration of troubleshooting, a parameter in our proposed method, that we even left out in our final tests.
In our paper, we describe that further machine learning techniques will be necessary for opimisation of the prioritisation method.
We used and will use expert knowledge to determine the relevant parameters that influence the criticality of disruptions and their possible weights. To verify and concretise these rather vague estimations, we will use more machine learning techniques.
If you have any further questions, do not hesitate to contact us.
Kind regards,
Jan Cetric Wagner and Gesa Wimberg
Dear Mr. Wagner, dear Ms. Wimberg,
thank you for your fast reply and useful indications.
Regards.
Roberto Teti