| Summary: | 碩士 === 中原大學 === 工業工程研究所 === 81 === In a manufacturing system that uses a robot as a material
handling device, deadlocking is defined as a situation in
which continuous part flow is inhibited because of a
conflict in part flow direction. Once a system deadlock
occurs, not only the machines and parts involved in the
deadlock will stop working, the entire manufacturing system
also loses its production capability. The presence of
system deadlocking is an obstacle in design and real
time control of flexible manufacturing systems. In a
flexible manufacturing system designed to have in-process
storages, deadlocking may be resolved with a detection and
recovery procedures. Firstly, the detection system
identifies the parts and machines participated in a
deadlock. Then the recovery system selects and transfers one
of the deadlocking part to a in-process storage, and the rest
of parts in the deadlock can proceed to its next machine.
Therefore, a system deadlock is resolved with appropriate use
of in-process storages. To install in-process storages in a
manufacturing syatem helps resolving system deadlock but also
increases the costs associated with the extra inventory.
Therefore to optimize the capacity of in-process storages is
necessary. This research develops a series of simulation
experiments to investigate the optimal capacity of in-
process storage in a manufacturing system using detection
and recovery method to resolve deadlocking. The
experimental results show that the maximun capacity of in-
process storage in a manufacturing system with M machines
is [M/2]- . When the routing of parts in a
manufacturing system is complicated, the occurrence of
deadlocking is frequent, and the utilization of in-process
storages is high.
|
|---|
