A novel approach for parallel disassembly design based on a hybrid fuzzy-time model
J Zhejiang Univ-Sci A (Appl Phys & Eng)
1673-565X
A novel approach for parallel disassembly design based on a hybrid fuzzy-time model*
Zhi-feng ZHANG 0
Yi-xiong FENG 0
Jian-rong TAN 0
Wei-qiang JIA 0
Guo-dong YI 0
0 (The State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University , Hangzhou 310027 , China)
1 Project supported by the National Natural Science Foundation of China (No. 51322506), the Zhejiang Provincial Natural Science Foundation of China (No. LR14E050003), the National High-Tech R&D Program (863 Program) of China (No. 2013AA041304), the National Key Technology Support Program (No. 2013BAF02B10), the Fundamental Research Funds for the Central Universities, and the Innovation Foundation of the State Key Laboratory of Fluid Power Transmission and Control , China ORCID: Zhi-feng ZHANG
This paper investigates the problem of parallel disassembly with the consideration of fuzziness. A novel approach is proposed based on optimized dispatching for parallel disassembly in which disassembly time is characterized by the fuzzy sets due to inevitable uncertainties. The proposed approach consists of three parts: in the first part, the fuzzy time-based dispatching disassembly process model is established; in the second part, the boundary conditions of the fuzzy time and the disassembly are derived, and the components' disassembly order and available stations are encoded together to find the optimal disassembly path; in the final part, the approach is optimized by using genetic algorithm (GA) to minimize the total time and cost, and the solution is compared with other algorithms. Finally, a case study for a hydraulic press disassembly is presented to verify the effectiveness and feasibility of the proposed approach.
Disassembly sequence planning; Dispatching disassembly process; Fuzzy time; Genetic algorithm (GA) doi; 10; 1631/jzus; A1500155 Document code; A CLC number; TB491
1 Introduction
Environmental problem has been a serious
challenge to our society, and many measures have been
promoted to ease this issue, such as public opinions,
industrial standards, and environmental laws
(Tian et
al., 2014)
. Disassembly, an important step of product
recycle, involves methodical extraction that departs
reusable components and valuable materials from
end-of-life products via certain operations.
There exist many research studies on
disassembly. Most researchers focus on three relative issues:
disassembly modeling, disassembly path planning,
and disassembly evaluation. Disassembly modeling
is the logical and informational basis for disassembly
that contains physics data, assembly data, and
constraint data. Disassembly path planning involves a
sequence of components removed from product
ontology. Disassembly evaluation is to assess the
disassembly scheme during the design to ensure
disassembility and efficiency of the products. Over the
decades, the issue on disassembly path planning has
attracted the attention of many researchers because it
relies not only greatly on environmental factors but
also on the efficient use of natural resources.
A number of researchers have proposed
disassembly sequencing approaches.
Kang et al. (2001)
proposed an approach based on an integer
programming formulation and considered
sequencedependent operation times to obtain the optimal
disassembly sequence.
Li et al. (2005)
presented
an object-oriented disassembly sequence for
maintenance based on the disassembly constraint
graph and the genetic algorithm (GA). González and
Adenso-Díaz (2006) adopted the scatter search
metaheuristic to deal with the disassembly cost of
complex products, when only one component can be
released at each time.
Lambert (2006)
modified the
two-commodity network flow approach to address
complex disassembly problems and to find out the
exact solutions, particularly, useful for evaluating
heuristic and metaheuristic approaches.
Aguinaga et
al. (2008)
analyzed the differences between the
disassembly path-planning problem and the general one,
and modified the rapidly-growing random tree-based
(RRT) algorithm to address these differences.
Tang
(2009)
introduced a fuzzy Petri net model to
explicitly represent the dynamics inherent in disassembly.
Li
et al. (2013)
proposed a modular disassembly
method to solve the inflexibility in a single modular of
complex mechatronics products.
Aydemir-Karadag
and Turkbey (2013)
dealt with a stochastic
disassembly line balancing problem and proposed a new
GA for multi-objective optimization. The objectives
related to line balance and design costs were
optimized by using an AND/OR graph.
Rickli and
Camelio (2013)
proposed a GA based on
disassembly operation costs, recovery reprocessing costs,
revenues, and environmental impacts to optimize partial
disassembly sequences.
Tian et al. (2013)
presented
some chance constrained programming models for
disassembly cost from the perspective of stochastic
planning. Moreover, two hybrid intelligent
al (...truncated)