Logistical Positioning in a Turbulent Environment

Robust capacity allocation in dynamic production networks

“…This makes capacity planning in the long-run a challenging task. Capacity levels have to be adapted in order to manage perturbations and trends of key parameters [2]. In particular, reduced or unavailable production capacities or an increased demand might put the timely satisfaction of customer demand at risk [3] -asking for a robust production network engineering.…”

“…In this context a robust production network engineering is given by the capability of the network to cope with several possible future scenarios in an efficient manner [5,6]. Planning approaches that rely on deterministic planning information [5] usually fail to cope with a dynamic environment and the considerable uncertainty of the underlying planning information [2]. Stochastic programming and robust optimization are two methods of setting up a robust plan that addresses uncertainty of relevant parameters [7].…”

Stability Analysis of Large Scale Networks of Autonomous Work Systems with Delays

“…Assumptions such as infinite capacity and fixed lead time are often made, leading to a static view of the production system may not be valid because WIP affects lead time and performance, while capacity is finite and varies both according to plan and due to unplanned disturbances such as equipment breakdowns, worker illness, market changes etc. Understanding the dynamic na-ture of production systems requires new approaches for the design of PPC based on company's logistics [16]. The controllers implicitly interact to adjust capacity to eliminate backlog as the system maintains its planned WIP level [15].…”

Fast and Reliable Order Management Design Using a Qualitative Approach

“…The definition of turbulence in OM follows physics [5]: Turbulence exists if individual values (microscopic view) deviate significantly from the representative value (macroscopic view of an aggregated value) [6]. To detect a significant deviation, it takes tolerances defining allowed deviations between individual values and the representative, i.e.…”

“…These conditions limit the prediction of an individual value by a mean-value based planning approach for reaching the required reliability. As a result, controlling turbulence takes two views [6,7]: • Objectively, turbulence can be identified by significant deviations -i.e. there are individual values deviating more than allowed from the mean value, Fig.…”