Journal of Materials Processing Technology 118 (2001) 403±410
Design and analysis of a virtual factory layout M. Iqbal*, M.S.J. Hashmi School of Mechanical and Manufacturing Engineering, Dublin City University, Dublin, Ireland
Abstract Conventional analytical systems are capable of providing statistical evaluation of targeted factory design feasibility. However, with this system, designers cannot get the feel for the actual setting of that factory design. With the advent of computer-aided design (CAD) system these problems are solved to a certain extent. CAD software allows the user to build models/templates and move them around on the screen and the analysis can be done in few hours. Re-layout could also be easily accomplished. But the drawback of the computer-based system is that they provide the user with a two-dimensional (2D) view of the layouts, which are not easy to visualise, understand and evaluate. With the virtual reality (VR) environment, a designer can have the feel of the actual setting of the factory. Inside the 3D environment, the depth perspective could be achieved which is not possible in a 2D view. This also provides better interactions than 2D system. Potential problems such as safety issues, aisle and other layout problems can be visualised and modi®ed by applying plant layout problem solving techniques. Re-layout of the existing factory layout can be done until satisfactory result is obtained. This paper describes how a factory layout can be designed and analysed in 3D virtual environment by applying factory layout problem solving techniques. An alternative layout was achieved and a new aisle system was introduced. Resulting material handling distance from department to department was found to be less after implementing the new aisle system in the factory. # 2001 Published by Elsevier Science B.V. Keywords: Computer-aided design; Virtual reality; Factory-layout
1. Introduction Factory layout involves the selection and arrangement of machines and material handling path, material handling devices, resulting in reduction in cost and time involved in manufacturing a product. Many methods exist by which a layout could be designed and analysed. Some of the earliest methods involved drawing plan views of different arrangements and after a detailed discussion with all the departments in the company picking the best one out of them. The other method involved moving template/models (both 2D and 3D) on a grid until a satisfactory layout was achieved [1,2]. These methods were inaccurate in terms of the exact locations and accurate time or cost analysis could not be done, overall they were quite cumbersome and time consuming. With the advent of computer-aided design (CAD) systems these problems were solved to a certain extent. Software was developed which allowed the user to build models/templates and move them around on the screen. CAD software allowed the user to build models/templates and move them around on the screen and the analysis can be done in few hours [3]. Re* Corresponding author. Tel.: 353-1-700-5000; fax: 353-1-700-5345. E-mail address:
[email protected] (M. Iqbal).
0924-0136/01/$ ± see front matter # 2001 Published by Elsevier Science B.V. PII: S 0 9 2 4 - 0 1 3 6 ( 0 1 ) 0 0 9 0 8 - 6
layout could also be easily accomplished. Inside the 3D environment, the depth perspective could be achieved which is not possible in a 2D view. This also provides better interactions than 2D system. Potential problems such as safety issues, aisle and other layout problems can be visualised and modi®ed by applying plant layout problem solving techniques. Re-layout of the existing factory layout can be done until satisfactory result is obtained. This paper tries to illustrate how plant layout problem solving technique can be implemented in 3D virtual environment to aid an engineer to solve the existing factory layout problem. It also shows the bene®ts of the improved visualisation provided by the virtual environment through which the designer is able to optimise space requirements and evaluate the relationships among machines, aisles and departments. 2. Virtual environment construction This section provides experience of creating virtual objects and placing them in the virtual environment with associated real world properties in order to illustrate how models are created. The equipment like lathe and drilling and milling were created in 3D Studio Max (release 4.0). An
404
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
initial design is obtained based on a block layout. All production machines were created in AutoCAD (release 13) and then imported to 3D Studio Max using the DXF format. First the main boundary wall of the factory was put in place and then aisles were created. Columns were placed to generate the constraints that are faced in real life situations. Partition walls were placed. Then the workshop and production machines were located on the factory ¯oor. In the following discussion, we present a detailed case study using the 3D Studio Max software and plant layout problem solving techniques. 3. Identifying bottlenecks on the factory floor To identify bottlenecks in a virtual factory, the designer needs to view various sections of the factory. Visualising different sections of a factory ¯oor under consideration, the following bottlenecks were identi®ed: 1. Visualising the packaging and storage sections (Figs. 1 and 2) it can be quickly realised that the existing storage area of the end products in front of the warehouse entrance blocks movement of people and materials. As a result the designer can quickly relocate the storage area of the end products stored in wooden boxes. 2. Visualising the 3D picture and 2D plan (Figs. 1 and 3), it can be observed that there is no systematic flow path of materials among the four sections. This leads to excess
effort in handling of materials as workers have to carry materials in an unplanned flow path. 3. Machines like milling machine, surface grinding, double ended grinder, drill machine and lathe (Fig. 2) can be relocated to another suitable place in order to have more space for the production section. This space can be utilised for setting up new production machines in future. 4. No aisle system has been implemented in the existing factory. Modi®cation to the existing facilities: 1. Introducing aisle system in the factory. 2. Rearrangements of workshop or production machineries. 3. Rearrangement of packaging section. 4. Rearrangement of finished product in open space. Bene®t to the modi®cation: 1. New aisle system has reduced material carrying distance (Figs. 4 and 5). Table 1 shows the difference of material handling distance before and after implementing the aisle system. 2. Changing the layout of production machines in nipple, spoke and washer section gives more space for setting up new production machineries in future (Fig. 4). Relocating the machines and carrying out the analysis permits creation of new layout. The area of the store room for packaging material has been increased from
Fig. 1. Bird eye view of the packaging area.
Table 1 Difference in material handling distance before and after implementing the new layout system (m) Machine type
Distance from the wire house to the machine before implementing aisle system
Distance of the wire house to the machine after implementing aisle system
Nipple forming m/c 1 Nipple forming m/c 2 Washer forming machine
27.13 25.31 31.25
19.21 21.19 22.41/23.02
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
405
Fig. 2. Old layout without aisle.
17.77 to 23.72 m2, thus giving more space for material and worker movement and storage areas. 3. Using new aisle system collisions among trolleys and boxes can be avoided when transporting the end products and production related materials from one place to another. 4. Finished products are stored in an organised way (Fig. 4). 3.1. Application of analytical methods to obtain a good general layout This section describes an analytical method, including computer-based heuristics, which can help construct and improve the functional layouts. A key issue in constructing the facility layouts is how to evaluate the goodness of
alternative layouts. Suppose we have a ¯ow cost matrix where cij is the cost per unit distance of all ¯ows between work centres i and j. For any given layout, let dij be the distance between work centres i and j. The cost associated with the ¯ow between i and j is Cijdij and the total ¯ow cost (TFC) for the ¯ows among all the work centres is XX TFC Cij dij i
j
where the summations are taken over all work centre pairs [4]. The TFC provides a concise way to compare specific layouts. Our task is to find the spatial arrangement of the work centres that makes the TFC as small as possible. Here the dij variable is indirect, determined by the spatial arrangement of the work centres [4].
406
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
Fig. 3. Material flow diagram before modifying the existing layout.
3.1.1. The relationship diagram In this layout analysis, the ¯ow of materials is the dominant factor. The from±to chart also referred as a travel chart which is used to measure the material ¯ow from one department to another department per day. Distance-based scores are inherently cost-minimising approaches to layout planning. It is assumed that distances between activities should be minimised, especially on frequently travelled routes. Table 2 and Fig. 6 show that E has the largest ¯ow and the ¯ow between E and B is the largest (130). So location of B has to be close to E. Similarly F has the second largest ¯ow (120) and the ¯ow between E and F is the second largest. Thus F has to be close to E. Flow between C and E is 100. So C has to be near E. Similarly ¯ow between D and E is 100 and so the location of D has to be near E. Position
Table 2 Estimated number of workers and materials moving between departments per daya From A B C D E F G H a
To A
B
C
D
E
F
G
H
± ± ± ± ± ± ± ±
35 ± ± ± ± ± ± ±
20 15 ± ± ± ± ± ±
10 5 70 ± ± ± ± ±
15 130 100 95 ± ± ± ±
15 10 10 10 120
± 5 8 ± 5 10
± 8 5 ± 5 ± 5 ±
A: main store; B: spoke, nipple and washer forming section; C: rim forming and polishing section; D: polishing section; E: electro plating section; F: packaging section; G: wire house-1; H: wire house-2.
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
407
Fig. 4. New layout with aisle.
of A, G and H are to be located away from production areas. All the sections can be arranged in the following manner [4]. Closeness rank of the departments are shown as in Table 3.
Combining the data from number of times people carry materials per day between departments (Table 2) and the distance between departments is shown in Table 4. We obtain the material handling distance travelled per Table 4 Distance (m) between departments after implementing the aisle system
Table 3 Closeness ranking of departments
From
Rank
Department
1 2 3 4 5 6 7 8
C E F G D B A H
A B C D E F G H
To A
B
C
D
E
F
G
H
± ± ± ± ± ± ± ±
33.50 ± ± ± ± ± ± ±
37.70 32.85 ± ± ± ± ± ±
50.80 48.25 13.85 ± ± ± ± ±
42.00 31.55 47.95 37.37 ± ± ± ±
22.10 23.73 34.45 47.56 37.78 ± ± ±
14.33 41.92 30.33 44.05 42.76 33.53 ± ±
8.14 31.84 45.65 42.98 42.98 19.51 23.63 ±
408
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
Fig. 5. Material flow diagram in new layout after introducing aisle.
day for the new layout with aisle system as shown in Table 5. PP Now using the formula TFC i j Cij dij to ®nd the total daily travel with layout (Fig. 2) we get 24881.40 per-
son-metre. Table 6 shows distance (m) between departments before implementing the aisle system. Again combining the data from number of times people carrying materials per day between departments (Table 2)
Table 5 Material handling distance travelled per day for the new layout after implementing the aisle system From
To A
B
C
D
E
F
G
H
A B C D E F G H
± ± ± ± ± ± ± ±
1172.5 ± ± ± ± ± ± ±
754.00 492.75 ± ± ± ± ± ±
508.00 241.25 960.50 ± ± ± ± ±
630.00 4101.50 4795.00 3358.10 ± ± ± ±
331.50 237.30 344.50 475.60 4533.60 ± ± ±
14.33 209.60 242.56 44.05 213.80 335.30 ± ±
8.14 254.72 228.25 42.98 214.90 19.51 118.15 ±
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
409
Fig. 6. Activity relationship diagram. Table 6 Distance (m) between departments before changes From A B C D E F G H
To A
B
C
D
E
F
G
H
± ± ± ± ± ± ± ±
40.85 ± ± ± ± ± ± ±
38.26 32.85 ± ± ± ± ± ±
51.98 48.69 13.72 ± ± ± ± ±
48.47 31.55 47.95 37.78 ± ± ± ±
27.14 23.92 34.14 47.56 37.78 ± ± ±
14.32 41.92 30.33 44.05 42.75 33.53 ± ±
17.37 31.78 41.76 42.98 46.26 19.51 23.63 ±
Table 7 Material handling distance travelled per day for the new layout with aisle system From A B C D E F G H
To A
B
C
D
E
F
G
H
± ± ± ± ± ± ± ±
1429.75 ± ± ± ± ± ± ±
765.20 492.75 ± ± ± ± ± ±
519.80 243.45 960.40 ± ± ± ± ±
727.05 4101.50 4795.00 3584.35 ± ± ± ±
407.10 239.20 341.4 475.6 4533.6 ± ± ±
14.32 209.60 242.64 44.05 213.75 335.30 ± ±
17.37 254.24 208.80 42.98 231.30 19.51 118.15 ±
and the distance between departments (Table 6), we obtain the material handling distance travelled per day for the new layout with aisle system as shown in Table 7. Similarly for layout (Fig. 4) the daily travel value is 25568.16 person-metre. Comparing the TFC value of the two layouts, the TFC value of the ®rst one (new layout) is low. So the new layout is chosen as the better layout.
4. Conclusion Virtual factory layout helps in evaluating plant layout before actually building them and assists in avoiding the costs involved in doing a physical re-layout. It also allows the user to get a better perspective than what could be achieved in 2D solutions. Re-location of the machines
410
M. Iqbal, M.S.J. Hashmi / Journal of Materials Processing Technology 118 (2001) 403±410
can be done such that the material handling cost is reduced as well as the bottlenecks are removed. References [1] V.S. Sheth, Facilities Planning and Materials Handling Ð Methods and Requirements, Marcel Dekker, New York, 1995.
[2] R.W. James, P.A. Alcom, A Guide to Facilities Planning, PrenticeHall, Englewood Cliffs, NJ, 1983. [3] D.P. Sly, Winter Simulation of a Conference'96 on Asystematic Approach to Factory Layout and Design with FactoryPLAN, FactoryOPT, and FactoryFLOW, December 1996, pp. 417. [4] J.S. Martinich, Production and Operations Management: An Applied Modern Approach, Wiley, New York, 1997, pp. 417.