2010 International Conference on Science and Social Research (CSSR 2010), December 5-7, 2010, Kuala Lumpur, Malaysia Assimilating Quality Function Deployment (QFD) with QUEST Analysis for Facility Layout Redesign of Handwork Section Mas Alina Mohd Ali a, Abdul Rahman Omar a, Alias Mohd Saman a, Ibrahim Othman b, Isa Halim a Abdul Hadi a a Faculty of Mechanical Engineering, Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia. e-mail: usu_ina@yahoo.com b Faculty of Art and Design, Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia. e-mail: ibkasut@yahoo.com. Abstract Quality function deployment (QFD) is a powerful customer-oriented tool to facilitate the customer needs. In this paper, an attempt has been made to examine the applicability of Quality Function Deployment (QFD) as a suitable design management tool for improving facility layout design of small and medium manufacturing industries. For this purpose, a simple case study using QFD has been conducted where survey has been done to collect and verify the expectations of the customer/worker profile for designing the facility layout. This paper presents the integration between QFD approaches with simulation modelling within a facility layout of a Handwork Section; Miyazu (M) Sdn Bhd, located in Shah Alam, Selangor. The department is struggling to locate the incoming parts from the depot since the pickup and drop off buffer is not properly placed, which results to untidy appearance. Quest simulation analysis is found to be very powerful tool that can be used in the design phase. Simulation modelling activity is used in this study since it allows verification of the floor design without having to rearrange the actual physical layout. The case study findings demonstrated that QFD approach is possible to prioritize and implement the new layout solutions at the onset of the design stage before the actual layout implementation. Keywords Quality Function Deployment; Facility Design Layout; Simulation Modeling I. INTRODUCTION Facility layout design is a well known key factor related to improvement of productivity. The efficient design of a facility layout has been recognized as one of the most important issues in contemporary manufacturing systems. It is generally accepted that effective facilities planning can reduce material handling cost by at least 10 to 30 percent. A good placement of facilities contributes to the overall efficiency of operations and can reduce about 50 percent of the total operating expenses [2]. Hence, if facilities are arranged optimally, manufacturers can reduce product cost and significantly enhance their competitive position [1]. Previous research [1, 2] indicate that limited attention paid to facilities design layout is a significant shortcoming since a wrong choice of machines /workers position can increase the expected travel distance. With rapid changes in market requirements, it is necessary to update the layout accordingly in order to operate efficiently. Hence, facility layout is not only affecting the production planning problem, but also an important part of manufacturing execution. The necessity for worker-friendly ambiance has initiated ever since facilities planning has taken a whole new meaning in the past 10 years [2, 8]. Thus, facility layout that able to accommodate accelerate productivity as well as providing comfort and safety to workers has become essential. The aim of satisfying the worker s requirement in the workplace environment can be achieved by integrating Quality Function Deployment (QFD) so that an efficient and establish design layout is able to be incorporated [3]. Quality Function Deployment (QFD) as a well-known customer-oriented methodology is widely used to assist decision making in product design and development in various types of production. Determining how and to what extent characteristics of products or services are to be met, with a view to gaining higher level of overall customer satisfaction is a key success factor in product design and development. The purpose of this research is to study the possible application of quality function deployment (QFD) methods and its suitability in determining the optimum layout parameters of manufacturing facilities based on the workers (customers ) satisfaction. In order to prove the relationship between the simulation analysis and performance matrices, an empirical facility layout model is developed first to illustrate its current performance. With this, a possible monitoring mechanism can be developed to quickly find out the adjustment solution once any configuration change is detected. II. METODOLOGY A case study was conducted in a metal stamping company situated in Shah Alam, Malaysia. One of the sections in the company s production is the handwork area. The main activity at the section is to perform restoration on the products of metal stamping process lines. The manufacturing operation at the production 978-1-4244-8986-2/10/$26.00 2010 IEEE 985
facility is run on two staged shifts including 80 minutes of tea breaks and lunch breaks per shift. This section was struggling to locate the incoming parts from the depot since the pickup and drop off buffer do not have its proper placed, which results to untidy appearance. The section is 30 meters in length and 11.5 meters wide were occupied with toilet, praying room, supervisor s office and eight working tables, differ in arrangements and shapes. All workers which are located at the respective table work simultaneously regardless of the difference in the amount and shapes of incoming parts that need to be restored. The incoming parts from the depot arrive at the area by two or more fork-lifts. The distance between the handwork area and the depot is 8 meters. Since the only space available to put the buffer at each working table is only 1.6 meter squared, the forklift has to locate the arrival parts at the space in between the depot and the handwork area which resulted congestion at the entry-exit points. The study was carried out in three stages. These approach was undertaken which began with identifying the type of facility layout used by gathering information from workers directly involved with the production operation. A. Questionnaire Surveys The first phase of the study was to gather and analyze all historical data. The production plant was visited for duration of one month with the assistance of the supervisor in charged. Data pertaining current production flow, process operations, processing times, resources and plant layout were gathered using proper time study methods. The questions were derived from several factors and design issues which strongly dependent on specific features of manufacturing system. The factors includes facility shapes, facility dimensions, pick-up and drop-off locations, product variety and product quantity, material handling systems and floor layout [4]. The questions are distributed among the industrial workers to acquire their requirements regarding the design of the facility layout. The questions are divided into five sections namely; 1) personal details of respondent, 2) facility shapes and dimensions, 3) material handling systems, 4) floor layout and 5) product quantity and product variety. Personal details of worker such as years of working experience were recorded in the first section. The information provided is vital to ensure that the workers understand and familiar with their layout surrounding. B. House of Quality Analysis The second phase commence with the process of developing the customer portion of the matrix by analyzing the information attained from the interview sessions. Figure 1 [5] illustrates the outline of the House of Quality design undertaken in this study. Once all the workers requirements have been obtained, the relative importance is calculated by multiplying each score by the number of worker who assigned it. The next step in HoQ analysis was to identify the technical importance of engineering characteristics. Figure 1: Simple flow diagram of QFD Process [5] By using the correlation between the customers / workers need and the engineering characteristic it was possible to determine the strength of relationship and impact on the need. It was possible to identify and prioritize the most important requirements and needs for the new design of layout. Using the lower part of the House of Quality matrix, the level of importance of the design requirements was determined using the results of the technical importance and relative importance of the engineering characteristic. Figure 2: HoQ step-by-step specifications determination 986
In this study, the competitive analysis was evaluated based on the standards design requirement using Malaysian Standards ISO 11064-4:2007; Ergonomic Design of Control Centers Part 4: Layout and Dimensions of Workstations. To complement the information for the Target Values definition, the level of importance of the design requirements was determined using the results of the Technical Importance and the Relative Weight of the Technical Requirements was developed using the lower part of the House of Quality matrix. After obtaining the final results of the Relative Importance of the Technical Requirements, it was possible to implement the new layout solutions and design of the facility layout of the manufacturing company. The improvement performance of the design layout was evaluated by measuring the differences of the design index before and after application of QFD. C. QUEST Simulation Analysis The final stage of the study began with building a reference model using simulation modelling. For the purpose of this research study, QUEST simulation software has been chosen as a platform to simulate the current and new improved layout design. Simulation is used in this study since it allows verification of the floor design without having to rearrange the actual physical layout. In order to represent the actual operation condition, the reference model was simulated using several probability distributions such as uniform distribution and constant distribution that may simulate the model closely to the actual layout and production condition. Results from the as-is model are statically compared with outputs of the real-world system to validate the simulation model [6]. When doing this simulation study, certain constraints are considered such: 1. Each job will be processed at any particular machine only once. 2. One machine can only process one job at one time. No two activities requiring the same machine to execute at the same time. 3. Handling time is included in the machine process time. Transportation time is negligible because the distance from machines to machines are near. 4. Allocation of the job to machine is subject to the earliest availability of the job. The reference model was built according to the existing facility layout design, to simulate the actual production condition. The model layout was designed using QUEST simulation software (Figure 3). Each run of the simulation model starts with all the material handlings, (forklift) parked at different blocks of storage area (depot). Based on the shift patterns, the forklifts deliver the parts from the storage area to the handwork area by following the forklift path (yellow lines). The forklifts follow the path network back and forth between handwork area and depot whilst deal faithfully with the driving regulations. When the forklifts reach each work station, it will place the incoming parts at the space provided near to the working table. Once the process is finish, the parts will be placed on the respective buffer before being collect by the forklift. Figure 3a: Plant layout of the current Handwork Area s model Figure 3b: Layout measurement of the current Handwork Area s model Simulation always runs for some future anticipated situation, therefore, there will always going to be assumptions made regarding any unknown future condition [6]. The assumptions include those placed on the characteristics of the jobs, machines and processing times such as: i) All jobs are simultaneously available at the beginning of the planning period. ii) No job cancellation occurs. iii) iv) Due date is known in advance. Raw materials, tools, jigs, fixtures etc, are presented and released immediately when required. v) Figure of all parts are represented by the same display neglecting the types and measurements of the entity. Validations of the reference model were made based on the comparison between the actual production output obtained from the company and simulation output results. The simulation was thus replicated for one day production cycles. The output variations were tested by running the simulation model for several hours over a few 987
cycles. The actual process of designing the layout to achieve the expected output of the proposed new facility is a function of the size of the work area, the size of the products and the work-in-process inventory at various workstations. While the size of the handwork area is a constant parameter, the size of products and the work-inprocess inventory factors are variable parameters that are driven by the input process of the production system. For the purposed of this research study, an approval with an official consent letter indicating the necessity to adhere the companies regulations was obtained from the companies involved allowing the author to conduct the research study at their premises. Before conducting the study, the researchers have also obtained research approval regarding the procedures from the Research Ethics Committee of University Technology MARA. III. RESULTS AND DISCUSSIONS The construction and redesign of the layout was made based on the information gained using the QFD approach. Several factors have been determined in redesigning a new improved layout such as the working space area, horizontal distance between workspace, space between work area and material handling and the load/unload station located at each workstation. The location of fan was also included as one of the factors that need to be considered during the improvement implementation. In order to gain the actual operation condition, the current model was simulated using several probability distributions that may simulate the model closely to the actual layout and production output condition. The simulation runs for one day of production cycle which is 22.2 hours including 160 minutes break for 2 shifts. The output variations were tested by running the simulation model for several cycles. Figure 4 shows the reference model that was built to represent the actual production layout in the QUEST simulation environment. From here, the simulation model validity is tested. The model was built based on the present production operation condition. The circles labeled as A, B, C, D, E, F and G shown in Figure 4 indicate parts placed outside the handwork area, i.e. in between the handwork section and the storage area (depot). This is due to space limitation which results to an inappropriate arrangement of the incoming parts. This condition may interrupt the forklift path as well as the flow of production. The average waiting time per part at each pick up and drop off location is 11.5 minutes while the average cycle time at each refurbishing process is 14.5 minutes. The working table arrangement shown in circles labeled I and II provide discomfort to workers. Since the worker s position are facing the fan (attached on the side of the wall), hence every time they work on the surface grinding process, dust from the metals flies directly to the workers. In the long run, this will affect the worker s health. Furthermore, the location of the working table represented by circle labeled III is not suitable for forklift to move back and forth. This eventually lead to longer waiting time and longer cycle time for each part and process respectively. B A Figure 4: Current facility layout design of Handwork Area Figure 5 portrays the model in an improved layout arrangement that was created using QUEST simulation software. The redesign of the layout was made with reference from information recorded using QFD. The size of the praying room and the supervisor s office are reduced to the same width. However, this will not disturb the worker s doings since they only require little space to continue with their usual activities alternately. By resizing these rooms, more space is available to reposition each working table. This rearrangement was made based on the worker s comfort ability to move around the working table. With this new arrangement, the workers can avoid the metal dust from flying directly to the workers. This will lead to a healthier working environment. Additionally, the problems with the placement of incoming parts are also solved by having this extra space. Hence, there will be no disturbance along the forklift path as well as for the production flow. Despite the space of the praying room and the supervisor s office being reduced, the workers are provided with a new resting area. This area is made so that these workers can have a place to drink and have a snack during their short breaks and lunch hour. Figure 5a: New improved facility layout design of Handwork Area I C II D III E F G 988
Figure 5b: Comparison between the current and the improved work area arrangement Figure 6a: QUEST summary report for original layout design Figure 5c: The fan position before and after design improvement Figure 6a indicates the summary of QUEST simulation production output of the current facility layout design. The results of the analysis using simulation software show the amount of incoming parts at source that need to be send to handwork area and the amount of finished parts at each working table that has been sent to storage area in sink. By comparison, it is obvious that both of the amounts did not match. For instant, finished parts processed by work table 3 are only 100 units, which differ from the amount of 208 units of parts that has been send earlier for refurbish. This proves that bottleneck of 108 parts occur on buffer located at work table 1. Figure 6b shows the summary of QUEST simulation production output of the new redesign layout. The analysis of the results signifies an increase in the amount of finished parts processed by each working table. The result shows that by doing some modification towards the facility layout design can increased the number of productivity made by workers. This is demonstrated by the increased amount of finished parts at each work table. Figure 6b: QUEST summary report for new layout design Table 1 compares the production output between the facility layout design before and after QFD. The results of the analysis using simulation software show that by applying QFD approach will not only increase the company s overall volume of production, but also minimize the workers degree of uncomfortable as well as reduce the risk of injuries. 989
TABLE 1 Design characteristics of the facility layout before and after QFD. Technical index Average waiting time per part Average cycle time per machine Overall production output per day Current Layout Design 11.5 minutes 14.5 minutes After Redesigning Layout Improvement 7.7 minutes 33 % 10 minutes 31 % 2330 parts 2580 parts 10.7% IV. CONCLUSIONS With the emergence of new generation of facilities design at the beginning of the century, there is a need to configure different approach to Facility Layout Problems. The new improved layout created smooth production flow and showed a good example of how facility design can be improved. Quality Function Deployment opens a new direction in terms of method to use to solve facility layout problems. The interrelationship is the most important mechanism of QFD process to stores valuable evidence needed for design improvements [7]. The use of simulation within layout can be an iteration process. The outcomes of the simulation model can be used as benchmarks for evaluating the efficiency of the new layout. ACKNOWLEDGMENT The author would like to acknowledge Miyazu (M) Sdn Bhd for the permission and ample opportunity to facilitate a fruitful case study. A special thank to Mr. Norazli Mohd, Supervisor of Handwork Area in Miyazu (M) Sdn Bhd for the assistance rendered during the related analytical work. Similar gratitude also goes to Faculty of Mechanical Engineering of University Technology MARA for providing facilities in carrying out this study. REFERENCES [1] S.S. Heragu, Facilities Design, PWS Publishing, Boston, 1997 [2] J.A. Tompkins, J.A. White, Y.A. Bozer, E.H. Frazelle, J.M. Tanchoco, and J. Trevino, Facilities Planning, 3 rd Edition, Wiley, New York, 2003 [3] M.M. Ali, A.R. Omar, A. Saman, I. Halim,, Improving facility layout using quality function deployment (QFD) approach Paper presented at the International Conference for Technical Postgraduates, Kuala Lumpur, Malaysia, ISBN:978-983-42035-9- 7, 14-15 December 2009. [4] A. Drira, H. Pierreval, and S.H. Gabouj, Facility Layout Problems: A survey Annual Reviews in Control, 2007, Vol. 31, pp. 255-267. [5] A.R. Omar, J.A Harding, and K. Popplewell, Design for Customer Satisfaction: An Information Modelling Approach The International Journal of Manufacturing Technology Management, 1999, Vol. 10, no 4, pp. 199-209. [6] C. Harrell, B.K Ghosh, and R. Bowden, Simulation Using ProModel, McGraw-Hill International Edition, United States of America, 2000 [7] M.M. Ali, and A.R. Omar, Integrating QFD and Simulation Analysis in Facility Layout Design Paper submitted to International Journal of Production Research for review. [8] F.E. Meyer & M.P. Stephens, Manufacturing facilities design and material handling, 3 rd Edition, Englewood Cliffs, NJ: Prentice Hall, 2005 990