TY  - CHAP
U1  - Konferenzveröffentlichung
A1  - Schuhmacher, Jan
A1  - Hummel, Vera
T1  - Training module on probabilistic forecasting of production and intralogistics turbulences within learning factories
T2  - Proceedings of the 12th Conference on Learning Factories (CLF 2022), 11-13 April 2022, Singapore
N2  - Industrial practice is characterized by random events, also referred to as internal and external turbulences, which disturb the target-oriented planning and execution of production and logistics processes. Methods of probabilistic forecasting, in contrast to single value predictions, allow an estimation of the probability of various future outcomes of a random variable in the form of a probability density function instead of predicting the probability of a specific single outcome. Probabilistic forecasting methods, which are embedded into the analytics process to gain insights for the future based on historical data, therefore offer great potential for incorporating uncertainty into planning and control in industrial environments. In order to familiarize students with these potentials, a training module on the application of probabilistic forecasting methods in production and intralogistics was developed in the learning factory 'Werk150' of the ESB Business School (Reutlingen University). The theoretical introduction to the topic of analytics, probabilistic forecasting methods and the transition to the application domain of intralogistics is done based on examples from other disciplines such as weather forecasting and energy consumption forecasting. In addition, data sets of the learning factory are used to familiarize the students with the steps of the analytics process in a practice-oriented manner. After this, the students are given the task of identifying the influencing factors and required information to capture intralogistics turbulences based on defined turbulence scenarios (e.g. failure of a logistical resource) in the learning factory. Within practical production scenario runs, the students apply probabilistic forecasting using and comparing different probabilistic forecasting methods. The graduate training module allows the students to experience the potentials of using probabilistic forecasting methods to improve production and intralogistics processes in context with turbulences and to build up corresponding professional and methodological competencies.
KW  - intralogistics
KW  - probabilistic forecasting
KW  - turbulences
KW  - analytics
KW  - machine learning
Y1  - 2022
UN  - https://nbn-resolving.org/urn:nbn:de:bsz:rt2-opus4-40835
U6  - https://doi.org/10.2139/ssrn.4073580
DO  - https://doi.org/10.2139/ssrn.4073580
SP  - 1
EP  - 6
S1  - 6
PB  - Elsevier
CY  - Rochester, NY
ER  -