Abstract:
On-demand digital-print service offers mass customization
and exemplifies personalized manufacturing services.
We describe a real-time and online optimization technique
based on genetic algorithms (GA) for print factory workflow
optimization. We have simulated digital-print factory manufacturing
activities as a heterogeneous, concurrent and integrated
system. The simulation is based on a virtual print factory,
which incorporates real factory characteristics such as successive
order acceptances, diverse production lines, various resource
types and quantities, and stochastic machine malfunctions. The
optimization objective is to reduce the number of orders that
miss deadlines, balance resource utilization, and ensure just-in time
production. The optimization technique has been integrated
into the virtual factory as a factory scheduler and resource assignment
engine. Significant improvements have been achieved
using the GA heuristic compared to baseline methods that are
currently implemented in an actual industrial setting.
