@article{9356,
  abstract     = {{In today’s manufacturing industry, enterprise-resource-planning (ERP) systems reach their limit when planning and scheduling production subject to multiple objectives and constraints. Advanced planning and scheduling (APS) systems provide these capabilities and are an extension for ERP systems. However, when integrating an APS and ERP system, the ERP data frequently lacks quality, hindering the APS system from working as required. This paper introduces a data quality (DQ) assessment framework that employs a Bayesian Network (BN) to perform quick DQ assessments based on expert interviews and DQ measurements with actual ERP data. We explain the BN’s functionality, design, and validation and show how using the perceived DQ of experts and a semi-supervised learning algorithm improves the BN’s predictions over time. We discuss applying our framework in an APS system implementation project involving an APS system provider and a medium-sized manufacturer of hydraulic cylinders. Despite considering the DQ assessment framework in such a specific context, it is not restricted to a particular domain. We close by discussing the framework’s limits, particularly the BN as a DQ assessment methodology and future works to improve its performance.}},
  author       = {{Herrmann, Jan-Phillip and Tackenberg, Sven and Padoano, Elio and Hartlief, Jörg and  Rautenstengel, Jens and Loeser, Christine and Böhme, Jörg }},
  issn         = {{1877-0509 }},
  journal      = {{Procedia Computer Science}},
  keywords     = {{Data Quality Assessment, Advanced Planning, Scheduling, Bayesian Network, Enterprise Resource Planning}},
  pages        = {{194--204}},
  publisher    = {{Elsevier}},
  title        = {{{An ERP Data Quality Assessment Framework for the Implementation of an APS system using Bayesian Networks}}},
  doi          = {{https://doi.org/10.1016/j.procs.2022.01.218}},
  volume       = {{200}},
  year         = {{2022}},
}

@misc{12803,
  abstract     = {{The increasing amount of alarms and information for an operator in a modern plant becomes a significant safety risk. Although the notifications are a valuable support, they also lead to the curse of overloading with information for the operator. Due to the huge amount of alarms it is almost impossible to separate the crucial information from the insignificant ones. Therefore, new procedures are required to reduce these alarm floods and support the operator to minimize the safety risk. One approach is based on learning a causal model that represents the relationships between the alarms. This allows alarm sequences that are causally implied to be reduced to the root cause alarm. Fundamental element of this approach is the causal model. Therefore in this work, different probabilistic graphical models are considered and evaluated on the basis of appropriate criteria. A real use case of a bottle filling module serves as a benchmark for how well they are suitable as a causal model for the application in alarm flood reduction.}},
  author       = {{Wunderlich, Paul and Hranisavljevic, Nemanja}},
  booktitle    = {{2019 IEEE 17th International Conference on Industrial Informatics (INDIN)}},
  isbn         = {{978-1-7281-2928-0}},
  keywords     = {{probabilistic graphical model, causal model, alarm flood reduction, Bayesian network, Markov chain, restricted boltzmann machine, automata}},
  location     = {{Helsinki, Finland }},
  pages        = {{1285--1290}},
  publisher    = {{IEEE}},
  title        = {{{Comparison of Different Probabilistic Graphical Models as Causal Models in Alarm Flood Reduction}}},
  doi          = {{10.1109/indin41052.2019.8972251}},
  year         = {{2019}},
}