A Batch Scheduling Model for a Three-stage Hybrid Flowshop Producing Products with Hierarchical Assembly Structures
This paper addresses a batch scheduling problem for a three-stage hybrid flowshop consisting of a machining stage processing common and unique components on unrelated parallel machines, an assembly stage combining the components into assembled products with complex assembly structures, and a dif...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Universitas Indonesia
2020-07-01
|
Series: | International Journal of Technology |
Subjects: | |
Online Access: | https://ijtech.eng.ui.ac.id/article/view/3555 |
Summary: | This paper addresses a batch scheduling problem for a three-stage hybrid
flowshop consisting of a machining stage processing common and unique
components on unrelated parallel machines, an assembly stage combining the
components into assembled products with complex assembly structures, and a
differentiation stage processing the assembled products on dedicated machines
to produce different product types. The common
components are the same for all products and are processed in batches, while
the unique components are dedicated to respective given product types and are
processed individually (one-by-one component). The goal is to schedule all
the products with different assembly structures to minimize total actual flow
time (TAFT) defined as total time
interval of components to be processed from their arrival times to their common
due date. A non-linear programming model is proposed, where small size problems
can be solved optimally using the LINGO software, and large size problems is to
be solved using a heuristic algorithm. The proposed algorithm consists of two
sub-algorithms. The first one is constructed using a shortest processing time
(SPT) based heuristic to get a job
sequence as an initial solution and the second one is to improve the initial
solution using the variable neighbourhood descent (VND) method with
neighbourhood insert and swap move operators. In solving the problem with the
algorithm, two scenarios arise, e.g., the same and the different sequences for
all stages. A set of hypothetical data is generated for different hierarchical
assembly structures to test the model and the algorithm, and the results show
that the different sequences for all stages obtain solutions with better
performances than the same ones. |
---|---|
ISSN: | 2086-9614 2087-2100 |