Integrating Model-based Diagnosis and Prognosis in Autonomous Production



Published Mar 26, 2021
Paul Maier Martin Sachenbacher Thomas Ru ̈hr Lukas Kuhn


Today’s complex production systems allow to si- multaneously build different products following individual production plans. Such plans may fail due to component faults or unforeseen behavior, resulting in flawed products. In this paper, we propose a method to integrate diagnosis with plan assessment to prevent plan failure, and to gain di- agnostic information when needed. In our setting, plans are generated from a planner before being executed on the system. If the underlying sys- tem drifts due to component faults or unforeseen behavior, plans that are ready for execution or already being executed are uncertain to succeed or fail. Therefore, our approach tracks plan exe- cution using probabilistic hierarchical constraint automata (PHCA) models of the system. This allows to explain past system behavior, such as observed discrepancies, while at the same time it can be used to predict a plan’s remaining chance of success or failure. We propose a formulation of this combined diagnosis/assessment problem as a constraint optimization problem, and present a fast solution algorithm that estimates success or failure probabilities by considering only a limited number k of system trajectories.

How to Cite

Maier, P., Sachenbacher , M. ., Ru ̈hr T. ., & Kuhn, L. (2021). Integrating Model-based Diagnosis and Prognosis in Autonomous Production. Annual Conference of the PHM Society, 1(1). Retrieved from
Abstract 6 | PDF Downloads 1



diagnosis, model based diagnostics, optimization, self-diagnosis

(Beetz et al., 2007) Michael Beetz, Martin Buss, and Dirk Wollherr. Cognitive technical systems — what is the role of artificial intelligence? In Proc. KI- 2007, pages 19–42, 2007.
(Beetz, 2000) Michael Beetz. Concurrent Reactive Plans: Anticipating and Forestalling Execution Failures, volume 1772 of Lecture Notes in Artificial Intelligence. Springer Publishers, 2000.
(Bouveret et al., 2004) S. Bouveret, F. Heras, Givry, J. Larrosa, M. Sanchez, and T. Schiex. Toolbar: a state-of-the-art platform for wcsp., 2004.
(Kask and Dechter, 1999) Kalev Kask and Rina Dechter. Mini-bucket heuristics for improved search. In Proc. UAI-1999, pages 314–32, 1999.
(Krause and Guestrin, 2007) Andreas Krause and Carlos Guestrin. Near-optimal observation se- lection using submodular functions. In Proc. AAAI-2007, pages 1650–1654. AAAI Press, 2007.
(Kuhn et al., 2008) Lukas Kuhn, Bob Price, Johan de Kleer, Minh Binh Do, and Rong Zhou. Pervasive diagnosis: The integration of diagnostic goals into production plans. In Proc. AAAI-2008, 2008.
(Kurien and Nayak, 2000) James Kurien and P. Pan- durang Nayak. Back to the future for consistency- based trajectory tracking. In Proc. AAAI-2000, pages 370–377, 2000.
(Mahtab et al., 2004) Tazeen Mahtab, Greg Sullivan, and Brian C. Williams. Automated Verification of Model-based Programs Under Uncertainty. In Pro- ceedings 4th International Conference on Intelli- gent Systems Design and Application, 2004.
(McDermott, 1993) Drew McDermott. A reactive plan language. Technical report, Yale University, Computer Science Dept., 1993.
(Mikaelian et al., 2005) Tsoline Mikaelian, Brian C. Williams, and Martin Sachenbacher. Model- based Monitoring and Diagnosis of Systems with Software-Extended Behavior. In Proc. AAAI-05, 2005.
(Schiex et al., 1995) Thomas Schiex, He ́le`ne Fargier, and Gerard Verfaillie. Valued constraint satisfaction problems: hard and easy problems. In Proc. IJCAI- 1995, 1995.
(Williams et al., 2001) Brian C. Williams, Seung Chung, and Vineet Gupta.Mode estimation of model-based programs monitoring system with complex behavior.In Proc. IJCAI-2001, pages 579–590, 2001.
Technical Papers