Akademik Veri Yönetim Sistemi
International Symposium on Computing and Artificial Intelligence (ISCAI 2022), Beijing, Çin, 16 - 18 Aralık 2022, ss.1
It is hard to find an optimization problem with
static parameters in real life industrial applications. Unpredictable machine
failures, unplanned/urgent job arrivals, job cancellations, unexpected changes
in production constraints, frequent changes in production lot sizes, changes in
delivery times, several changes in short-mid-term plans such as job-work assignment,
production planning are some well-known repeatedly encountered dynamic events
in actual production systems. One of the main characteristic of these dynamic events
is that, they cannot be known precisely beforehand and cannot be predicted
easily. The problem domains that change over time (time-varying) or
change via some unexpected events (event-based) or time-varying variants
are usually called as dynamic environments
and the problems with these characteristics are called to as dynamic optimization problems (DOPs). In
this invited talk, several real life industrial applications of DOPs
from our previous studies will be presented. These include; dynamic scheduling
of heat treatment furnaces, dynamic part family formation for cellular
manufacturing, dynamic scheduling of flexible manufacturing systems with
flexible transportation abilities, dynamic load consolation for transportation
operations, dynamic optimization of several CNC operations etc.