Adjustment mode decision based on support vector data description and evidence theory for assembly lines

Pages1711-1726
DOIhttps://doi.org/10.1108/IMDS-01-2017-0014
Publication Date10 Sep 2018
AuthorYoulong Lv,Wei Qin,Jungang Yang,Jie Zhang
SubjectInformation & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems
Adjustment mode decision
based on support vector data
description and evidence theory
for assembly lines
Youlong Lv
College of Mechanical Engineering, Donghua University, Shanghai, China
Wei Qin and Jungang Yang
Shanghai Jiao Tong University, Shanghai, China, and
Jie Zhang
College of Mechanical Engineering, Donghua University, Shanghai, China
Abstract
Purpose Three adjustment modes are alternatives for mixed-model assembly lines (MMALs) to improve
their production plans according to constantly changing customer requirements. The purpose of this paper is
to deal with the decision-making problem between these modes by proposing a novel multi-classification
method. This method recommends appropriate adjustment modes for the assembly lines faced with different
customer orders through machine learning from historical data.
Design/methodology/approach The decision-making method uses the classification model composed of an
input layer, two intermediate layers and an output layer. The input layer describes the assembly line in a
knowledge-intensive manner by presenting the impact degrees of production parameters on line performances.
The first intermediate layer provides the support vector data description (SVDD) of each adjustment mode through
historical data training. The second intermediate layer employs the DempsterShafer (DS) theory to combine the
posterior classification possibilities generated from different SVDDs. The output layer gives the adjustment
mode with the maximum posterior possibility as the classification result according to Bayesian decision theory.
Findings The proposed method achieves higher classification accuracies than the support vector machine
methods and the traditional SVDD method in the numerical test consisting of data sets from the machine-
learning repository and the case study of a diesel engine assembly line.
Practical implications This researchrecommends appropriateadjustment modes for MMALs in response
to customerdemand changes. Accordingto the suggested adjustmentmode, the managers can improvethe line
performance moreeffectively by using the well-designedoptimization methods for a specificscope.
Originality/value The adjustment mode decision belongs to the multi-classification problem featured with
limited historical data. Although traditional SVDD methods can solve these problems by providing the
posterior possibility of each classification result, they might have poor classification accuracies owing to the
conflicts and uncertainties of these possibilities. This paper develops a novel classification model that
integrates the SVDD method with the DS theory. By handling the conflicts and uncertainties appropriately,
this model achieves higher classification accuracies than traditional methods.
Keywords Decision making, Adjustment modes, DempsterShafer theory, Mixed-model assembly lines,
Support vector data description
Paper type Research paper
1. Introduction
Mixed-model assembly lines (MMALs) are widely used in manufacturing industries because
they can assemble various customized products in an intermixed sequence with less setup
time (Becker and Scholl, 2006; Caputo and Pelagagge, 2008). In the MMAL, production
configurations such as the workforces and operation contents at each station are balanced Industrial Management & Data
Systems
Vol. 118 No. 8, 2018
pp. 1711-1726
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-01-2017-0014
Received 22 November 2017
Revised 26 February 2018
Accepted 1 April 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
The work described in this paper was supported by the Fundamental Research Funds for the Central
Universities (No. 2232018D3-28), Shanghai Sailing Program (No. 18YF1400800) and the National
Natural Science Foundation of China (No. 51435009; No. U1637211).
1711
Adjustment
mode decision
for specific model mixes at the design stage (Emde et al., 2010). This might lead to the
production imbalance whenever the demand structure from customer orders shifts to
another model mix because different product models often utilize the available capacity in
very different intensities (Caputo et al., 2015; Lian et al., 2012). Therefore, production
adjustments are important for MMALs to keep their performance at high levels, especially
in constantly changing production environments.
According to the production planning framework of MMALs generalized by Boysen et al.
(2009a, b), two adjustment scopes are alternatives in response to new customer orders, i.e. the
configuration scope and the scheduling scope. The balancing problem deals with
the adjustments in the configuration scope for new productivity, which is defined as the
reconfiguration mode. On the contrast, the sequencing problem modifies the scheduling scope
based on new demand structures, known as the rescheduling scope. Both problems have been
widely investigated by scholars with the considerations of various assumptions and optimization
targets (Akpinar and Bayhan, 2014; Cortez and Costa, 2014; Shao et al., 2010). In addition, some
studies have dealt with the balancing and sequencing problem in an integrated manner to realize
the global optimization, although its increased complexity causes more computational efforts
(Hamzadayi and Yildiz, 2013; Saif et al., 2014). In general, MMALs depend on the rescheduling
mode, the reconfiguration mode and the integrative mode to adjust their production in the
scheduling scope, the configuration scope and the overall scopes, respectively. Of these modes,
the integrative mode is the most suitable for performance improvement, but requires too much
computational efforts when the adjustment is effective in a single scope. The reconfiguration
mode or the rescheduling mode is useless when the production parameters of its related scope
have little effects on MMAL performances. Thereupon, adjustment mode decision is an essential
problem for MMALs to accommodate to the demand volatility from customer orders.
This problem is complicated because either the balancing problem or the sequencing
problem is featured with many constraints, conflicting objectives and numerous parameters
(Boysen et al., 2011). The comparative analysis between these modes requires too massive
simulation efforts and is very sensitive to subjective preferences. To our knowledge, the
research on adjustment mode decision is still in its infancy, and the practical application is
rare. Without the support of scientific decision-making methods, MMAL managers choose
the adjustment modes based on their own experience in most cases. To fill in this gap, this
paper proposes a new approach that integrates the support vector data description (SVDD)
method with the DempsterShafer (DS) theory to recommend appropriate adjustment
modes for different MMAL conditions. Owing to the essence that adjustment modes indeed
modify the production parameters of different scopes to improve MMAL performances, the
proposed approach uses the impact degrees of production parameters on performance
indices to represent MMAL conditions in a knowledge-intensive way. Based on the
historical data consisting of these impact degrees and related adjustment modes, the SVDD
method builds the posterior probability distribution of each mode through data training of
decision boundaries, while DS theory combines the probabilities of different SVDDs by
dealing with their uncertainties and conflicts. The adjustment mode with minimum decision
risk is finally recommended according to the combined probabilities.
The remainder of this paper is organized as follows. In Section 2, the background of
adjustment mode decision problem is addressed. Section 3 presents the SVDD-based
decision approach. Section 4 gives the comparative experiment between the proposed
method and other methods. Finally, conclusions are outlined in Section 5.
2. Background
2.1 Mixed-model assembly line adjustment modes
In MMALs, the production planning based on a new scheduling problem is most widely
used to deal with customer demand changes. This regularly starts from the master
1712
IMDS
118,8

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT