DISASSEMBLY SYSTEM DESIGN WITH OPTIMAL ENVIRONMENTAL AND ECONOMIC
PARTS SELECTION USING LIFE CYCLE INVENTORY DATABASE BY INPUT-OUTPUT
TABLES
Tetsuo Yamada*, The University of Electro-Communications
1-5-1 Chofugaoka, Chofu, Tokyo 182-8585, Japan, tyamada@uec.ac.jp, (+)81-42-443-5269
Kento Igarashi, The University of Electro-Communications, Tokyo, Japan
Norihiro Itsubo, Tokyo City University, Yokohama, Japan
Masato Inoue, Meiji University, Kawasaki, Japan
ABSTRACT
To prevent global warming by supply chain, End-of-Life (EOL) assembly products should be
disassembled environmentally and economically for material circulation (Wang and Gupta, 2011). This
paper proposes a disassembly system design with an optimal environmental and economic parts
selection which harmonizes collected CO2 volumes and recycling cost using a Life Cycle (LC)
inventory database by the input-output tables (Yamada et al., 2012). The first step is to optimize the
environmental and economic parts selection with the integer programming, and the second step is to
carried out the line balancing for minimizing the number of stations.
Keywords: Low-carbon and closed-loop supply chain, Recycling, Environmentally-conscious
manufacturing, Sustainable manufacturing, Integer Programming
INTRODUCTION
To prevent global warming by supply chains, End-of-Life (EOL) assembly products should be
disassembled not only environmentally but also economically for material circulation [1]. With
disassembly parts selection in recycling factories, parts/materials with higher CO2 volumes should be
disassembled for environment if the CO2 volumes of each part can be estimated. On the other hand,
ones with lower recycling cost should be also disassembled for economy if the recycling cost of each
part can be estimated. In addition, a disassembly line balancing should be also carried out based on the
optimal environmental and economic parts selection [2]. However, there is another design issue how to
obtain product and environmental information such as the disassembly time and CO2 volumes. To
overcome this issue, the Life Cycle (LC) inventory database by the input-output tables [3] and
Recyclability Evaluation Method (REM) software developed by Hitachi. Ltd. [4] can be used.
This paper proposes a disassembly system des ign with the optimal environmental and economic parts
selection which harmonizes collected CO2 volumes [3] and recycling cost [4] using the LC inventory
database by the input-output tables. The first step is to optimize the environmental and economic parts
selection with the integer programming [5], and the second step is to carried out the line balancing for
minimizing the number of stations with the ranked positional weight heuristic [6].
DISASSEMBLY DESIGN PROCEDURE WITH OPTIMAL ENVIRONMENTAL AND
ECONOMIC PARTS SELECTION USING LIFE CYCLE INVENTORY DATABASE BY
INPUT-OUTPUT TABLES
Disassembly System Design with Optimal Environmental and Economic Parts
Selection using Life Cycle Inventory Database by Input-Output Tables
Disassembly System Design
(Yamada & Sunanaga, 2011)
i) Construction and Analysis of Product
Recovery Values with BOM
Environmental and Economic Parts Selection
using Life Cycle Inventory Database by InputOutput Tables (Yamada et al., 2012)
a) Identification of Department Name in
Input-Output Table
b) Calculation of Unit Price for Each Part
ii)Maximal Cycle Time
c) Calculation of CO2 Volumes for Each Part
iii) Condition on the Number of Stations
iv) Line balancing with Environmental
Loads
v) Line Evaluation with Product Recovery
Values
d)Estimation of Recycling Cost using
Recyclability Evaluation Method (REM)
e) Environmental and Economic Parts
Selection by Integer Programming with ε
constraint
Environmental and Economic Parts Selection
using Recyclabiity Evaluation Method REM
(Igarashi et al., 2012)
Figure 1 Design procedure for a disassembly system with an optimal environmental and economic parts
selection using life cycle inventory database by input-output tables
This paper proposes a design procedure for a disassembly system with an optimal environmental and
economic parts selection using LC inventory database by the input-output tables as shown in Figure 1.
This design procedure consists of 2 main flows from upstream to downstream for the disassembly
system design on the left side [3] and the environmental and economic parts selection on the right side
[2]. In the environmental and economic parts selection, the LC inventory database by the input-output
tables with the bill of materials [2] are adopted to estimate the CO2 volumes for each part as the
environmental loads, while the Recyclability Evaluation Method (REM) software developed by Hitachi.
Ltd. [4] is used to estimate the disassembly times and recycling rates of each part.
The LC inventory database used in this study is calculated by the Japanese input-output tables [3]. In
general, the input-output tables define economic relationships among sectors by matrix representation
based on annual transactions among sectors, so that the carbon dioxide emission intensity is obtained by
using the LC inventory database by the input-output tables. With the LC inventory database by the I/O
tables, the CO2 volumes at each part are estimated with the product information such as prices and
weights [3]. On the other hand, the disassembly time and recycling cost of each part are estimated by
inputting product information such as material type, weight and disassembly motion at each part to the
REM software [4]. In the software, the recycling cost is the differences between the recovered material
prices and costs, where the costs consist of disassembly, material process and disposal costs,
respectively. If the recovered material prices are higher than the costs, the value of the recycling cost is
negative which means positive profits earned by the recycling.
FORMULATION OF OPTIMAL ENVIRONMENTAL AND ECONOMIC DISASSEMBLY
PARTS SELECTION
With the product disassembly data and CO2 volumes obtained by the LC inventory database [3] and the
REM [4], 0-1 integer programming [4] is used in this study for the selection of the parts disassembled or
not in terms of the CO2 volumes and the recycling cost similar to [2]. The combinatorial solution which
maximizes the collected CO2 volumes but minimizes the total recycling cost of the product is examined
to satisfy the constraints of the disassembly precedence relation. The notation of the disassembly parts
selection used for the integer programming is as follows:
cj
ej
E
Emax
C
N
xi
: Recycling cost at part j
: CO2 volumes at part j
: Total collected CO2 volumes at a product
: Maximal CO2 volumes of a product in all parts disassembled
: Total recycling cost at a product
: Number of parts
: Binary value; 1 if part i is disassembled, else 0
: Constraint of total CO2 volumes of selected parts
: An arc with constraints of disassembly precedence relation
: An arc without constraints of disassembly precedence relation
: set of tasks that immediately precede part j
ε
A1
A2
Pj
Similar to [2], the objective functions for minimizing total recycling cost and maximizing total CO2
volumes are respectively set as equations (1) and (2):
N
C   c j x j  Min
(1)
j 1
N
E   e j x j  Max
(2)
j 1
The constraint of the disassembly precedence relations are set as equations (3), (4) and (5) [2] based on
Nof et al. [6]:
xi  x j  0　i  Pj
Subject to:
(3)
A  A1  A2 (i, j )  A
(4)
To solve this multiple purpose optimization, ε-constraint method is used as well as [2]. The objective
function E is made into the only objective function, a nonlinear optimization is performed to each of
those combinations by changing ε gradually. The function E looks for the Pareto optimum solution set.
Then E is transposed to
E  .
(5)
DESIGN EXAMPLE OF DISASSEMBLY SYSTEM WITH OPTIMAL ENVIRONMENTAL
AND ECONOMIC PARTS SELECTION USING LCI DATABASE BY INPUT-OUTPUT
TABLES
To validate the proposed design procedure of the disassembly system, an example of the assembly
product is prepared. A cleaner is prepared as an example of 3D-CAD model [7]. The production plan is
also prepared as shown in Table 1.
To harmonize the environmental and economic aspects in the obtained disassembly part selection with
the integer programming [2][8], four scenarios as well as [2] are here considered and discussed for the
product evaluation as follows: 1) All parts disassembled, 2) CO2 volumes maximum, 3) CO2 volumes
and cost coexistence and 4) Recycling cost minimum. In the scenario 2) CO2 volumes maximum, a
solution with the highest value of the total collected CO2 volumes at the product, E, is selected within
the candidates whom their collected CO2 volumes is higher than 50 [%]. For the line evaluation, the
disassembly line balancing is carried out by the ranked positional weight heuristic [6] for the selected
disassembly parts at each scenario, respectively.
Table 1 Example of disassembly problem for cleaner
Demands Q for Collected EOL products
Production Planning Period T0
during T0
8,400 [min] (= 20 [days] × 7 [hours] × 60 [min])
12,000
450
400
350
Recycleing cost
300
250
200
150
100
Figure
2 Behaviors of Recycling Cost for CO2 Volumes
50
21.77
20.06
17.49
17.49
17.49
17.49
03.54 52.07
13.2
15
45500
10.5
14.52
42.66
35.68
29.93
45231.63
16.2
1
2
3
13.2
13.2
46000
5
4
13.37
13.37
17.49
51.25
14.15
53.47
10.2
10.2
13.2
6
7
194.97
17.49
15.6
17.49
186.04
50.81
357.15
13.2
10
Scenario 3: CO2
volumes and
cost coexistence
：Disassembly time [sec]
：Recycling cost index
：CO2 volume [g-CO2]
15.6
13
13.2
14
30.33
13.2
13.2
17.49
18.41
72.30
452.55
47500
32.61
13.2
St3
18
18.63
8.96
17.49
12.52
93.89
61.99
15.60
66.89
13.2
3.85
13.2
22
13.2
23
20
21
13.2
13.2
11
17
St2
St1
47000
13.2
19
CO217.49
volumes
[g-CO2]
17.49
435.13
16
27.6
9
18.37
8
36.51
46500
13.2
15
12
1. Wheel
2. Wheel stopper
3. Upper nozzle
4. Lower nozzle
5. Nozzle
6. Right handle
7. Switch
8. Left handle
9. Left body
10. Right body
11. Dust case cover
12. Mesh filter
13. Connection pipe
14. Dust case
15. Exhaust tube
16. Upper filter
17. Lower filter
18. Protection cap
19. Motor
20. Rubber of outer flame of fan
21. Outer flame of fan
22. Lower fan
23. Fan
Figure 3 Precedence relations among disassembly element tasks with optimal environmental and
economic parts selection: Scenario 3) CO2 volumes and cost coexistence
48000
scenario 3: CO2 volume and cost coexistence
42
36
16
20
12
18
11
17
19
st1
st2
st3
30
24
18
21
12
6
0
Figure 4 Pitch diagram with optimal environmental and economic parts selection:
Scenario 3) CO2 volumes and cost coexistence
Figure 2 shows the Pareto optimal solution for the recycling cost and CO2 volumes in the experiment.
While the collected CO2 volumes from the disassembled parts in one product ar e shown on the
horizontal axis, the recycling cost is shown on the vertical one, where each solution is obtained by each ε
constraint. Figure 3 shows the precedence relations among disassembly element tasks after the
environmental and economic parts selection in the scenario 2, CO2 volume maximum. “x” marks in the
figure means the canceled disassembly tasks with the non-selective parts. By using the disassembly
precedence relations among the selected tasks, the disassembly line balancing is carried out by t he
ranked positional weight heuristic [5]. The assignment of each task to stations are also shown in Figure 3,
and the pitch diagram with the optimal environmental and economic parts selection are drawn as shown
in Figure 4.
Table 2 Example of disassembly system design using LC inventory database with input-output tables
Constraints
Product
Evaluation
Line
Evaluation
Total disassembly time [sec]
Number of parts
Constraint of required CO2 volumes [%]
Actual CO2 volumes [g-CO2]
Recycling cost index
Minimal
Number of stations
Actual
Balance delay (BD)
Smoothness index (SI)
Scenario 1: All parts
disassmbled
316.2
23
100
47579.28
402.17
8
8
0.06
8.38
Scenario 2: CO2
Scenario 3: CO2 volumes Scenario 4: Recycling cost
volumes maximum
and cost coexistence
minimum
219
102.6
52.8
16
8
4
90
50
10
47347.30
45729.71
46410.44
272.89
127.34
63.85
5
3
2
5
3
2
0.18
0.07
0.37
24.98
18.91
26.40
Table 2 show an example of the disassembly system design using the LC inventory database with the
input-output tables. In the product evaluation, the total disassembly time and the number of
disassembled parts basically decrease in order to reduce the recycling cost from scenario 1 to 4, so that
the collected CO2 volumes also decrease as the recycling cost decreases. From the viewpoint of the CO2
volumes, there are a few differences between the scenarios within only 3.9%. However, the recycling
cost at the scenarios 2 to 4 is 1.5 to 6.3 times lower than one at the scenario 1. One of the reasons is that
a part “Motor” is the largest CO2 volumes among the all parts, therefore, the total CO2 volumes at the
product is almost maintained as long as a part with the largest CO2 volumes such as the motor is
selected and disassembled.
In the line evaluation, the disassembly line balancing is carried out by the ranked positional weight
heuristic [6] for the selected disassembly parts at each scenario, respectively. As the total disassembly
time and the number of disassembled parts basically decrease, the minimal and actual number of stations
is decreased.
SUMMARY AND FUTURE STUDIES
This paper proposed the disassembly system design with the optimal environmental and economic parts
selection which harmonized collected CO2 volumes and recycling cost using the LC inventory database
by the input-output tables and the recyclability evaluation method. The design example demonstrated
that the recycling cost was minimized in spite of maintaining the total collected CO2 volumes by
selecting a disassembled part with the largest CO2 volumes.
Further study should evaluate the CO2 volumes of each part by the process base inventory database, use
the LC inventory database by the input-output tables in the other countries, optimize the line balancing
under the optimal environmental and economic parts selection, etc.
ACNOLEDGEMENTWS
The authors would like to thank Mr. Takayuki Nishi and Hitachi, Ltd. for REM software. This research
was partially supported by the Environment Research and Technology Development Fund (E -1106) of
the Ministry of the Environment, Japan.
REFERENCES
[1] Wang, H.-F., Gupta, S. M., Green Supply Chain Management: Product Life Cycle Approach. New
York, NY: McGraw Hill, USA, 2011.
[2] Igarashi, K., Yamada, T., Inoue, M., “Disassembly System Design with Environmental and
Economic Parts Selection using Recyclability Evaluation Method”, Proceedings of the International
Symposium on Reliability Engineering and Risk Management (ISRERM2012) , Aug, Yokohama,
Japan, CDROM, 177-182, 2012
[3] Yamada, T., Suzuki, Y., Masui, T., Itsubo, N., Inoue, M., “Disassembly System Design with
Environmental and Economic Parts Selection using Life Cycle Inventory Database by Input -Output
Tables”. 2011, Oct, 10th Global Conference on Sustainable Manufacturing, Sustainable Product
Development and Life Cycle Engineering, CIRP, Istanbul, Turkey, 321-326.
[4] Recyclability
Evaluation
Method
(REM),
©
Hitachi,
Ltd.,
EcoAssist
:
http://www.ecoassist.com/HTML_n/option/rem/rem_tr/ppframe.htm
[5] Kubo, M., Combinatorial Optimization and Algorithm, Tokyo: Kyoritsu Shuppan, 2000 (in
Japanese).
[6] Nof, S.Y., Wilhelm, W.E. and Warnecke, H., Industrial Assembly, CHAPMAN & HALL, 1997.
[7] Inoue, M., Akiyama, T. and Ishikawa, H. “Lifecycle design support system based on 3D-CAD using
set-base design method,” The Japan Society of Mechanical Engineers, 21th Conference on Design
Engineering and System Department, Oct, Osaka, 420-423, 2011 (in Japanese).
[8] GLPK - GNU Project, http://www.gnu.org/software/glpk/ (Available on Jan 9, 2012)