(PDF) A Complex Garment Assembly Line Balancing Using Simulation-based Optimization

Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx RESEARCH ARTICLE A Complex Garment Assembly Line Balancing Using Simulation-based Optimization Ocident Bongomin*1 | Josphat Igadwa Mwasiagi1 | Eric Oyondi Nganyi1 | Ildephonse Nibikora2 1 Department of Manufacturing, Industrial and Textile Engineering, Moi University, Abstract Eldoret, Kenya 2 The nascent wave of disruptive competition in the Department of Polymer, Industrial and Textile Engineering, Busitema current business environment brought about by the University, P.O. Box 236, Tororo, fourth industrial revolution (Fashion 4.0 or Apparel Uganda 4.0) is enormous. Therefore, it is paramount important to apparel industry to be flexible enough to respond Correspondence quickly to the unstable customers’ demand through *Ocident Bongomin, Department of Manufacturing, Industrial and Textile continuous improvement of their process efficiency Engineering, Moi University, Eldoret, and productivity. This study aims at achieving an Kenya optimal trouser assembly line balancing using Email:

simulation-based optimization via design of experiment. The empirical study is conducted at Funding Information Southern Range Nyanza Limited (NYTIL) garment ACE II- PTRE Grant/Award Number: Credit No. 5798- facility and a complex trouser assembly line with 72 KE operations is considered. The discrete event simulation of the trouser assembly line is developed using Arena simulation software. The local optimal solution is obtained from simulation experimentation and is adopted for the optimization process. The OptQuest tool is utilized to solve a single objective function (throughput) optimization problem. The results show that average throughput increases from the existing design (490 pieces per day) to local optimal design (638) and global optimal design (762). Consequently, the line efficiency increases from 61.2% to 79.7% to 95.2% respectively. The high increase in line efficiency and average throughput confirms the suitability assembly line balancing using simulation-based optimization via design of experiment. KEYWORDS Arena, Discrete event simulation, Objective function, OptQuest, Optimization, Simulation modeling 1 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx 1 | INTRODUCTION The engineering marvels imputed by Fordism and Taylorism in the early age of the second industrial revolution led to the advent of assembly line [1]. It is unprecedented and often credited as one of the most significant developments in the modern world [2]. Assembly line systems typically include a high proportion of operations performed manually, which results in variable operating times for the same operation that contributes to its complex nature [3]. Besides, tasks are allocated to the workstations considering some restrictions including precedence constraints, cycle time and number of workstations and thus increasing its complexity [4]. However, assembly lines are used extensively in mass production systems to produce high quantity standardized products [5]. For this reason, assembly line design or balancing becomes very crucial for proper functioning of the assembly line system. The design or balancing of assembly lines is an important issue in manufacturing engineering, management and control [6–8]. However, assembly line balancing is a very complex phenomenon which has been dubbed as a NP (non-polynomial)-hard problem or complex combinatorial problem [9]. Assembly line balancing is the act of assigning the tasks to the workstations by optimizing the pre-specified objective function without violating the precedence constraints. Assembly line balancing that produces a single product model is referred to as a simple assembly line balancing problem [5, 10–12]. Despite the fact that research on assembly line balancing dated back for over a century, it is still of interest to vast researchers in the present days. This is because assembly line balancing problem is directly related to the production efficiency [13]. Moreover, garment assembly or sewing line exhibits a strange and complex balancing problem as it employs huge number of processes, machines and operators [14]. Therefore, optimal design would provide the best solution for this kind of problem. However, in the previous studies, garment assembly line balancing has been implemented using several techniques including manual/practical [15], ranked positional weight [16], COMSOAL [17], largest candidate rule [18], simulation [19], genetic algorithm [20] and hybrid (simulation and heuristics) [21]. As the complexity of garment assembly line increases depending on the fashion styles, most of these methods become less effective and inferior for designing it. Simulation-based optimization is the state-of-the-art design technique that combine both simulation and optimization technique. Application and development of this method is increasing drastically as it is one of the key disruptive technologies shaping the era of the fourth industrial revolution (industry 4.0) [22]. Previously, vast researchers have dealt with assembly line balancing using simulation and optimization methods albeit independently. Since the advent of computer technology, simulation has been commonly used in a wide spectrum of fields including manufacturing, healthcare, marketing, transportation and supply chain. Of all fields, manufacturing systems emerged to be the most important area in which simulation has been successfully adopted in numerous studies [23]. However, most practical manufacturing systems are extremely complex that finding optimal decision variables analytically are enormously difficult. Therefore, simulation-based optimization is widely used in evaluating complex systems and optimizing responses for manufacturing problems [24]. Several studies on garment assembly line balancing using simulation technique have been reported in the recent past. For instance, Kitaw et al. [25] developed an approach for assembly line balancing for garment production using Simul8 simulation software. Kursun and Kalaoglu [26] also conducted simulation study of production line balancing in apparel manufacturing using Enterprise-dynamics simulation software. Stief et al.[27] used plant simulation software, while Hanan & Seedahmed [28] employed MySQL Data Base Management System (DBMS), Java language, Hyper Text Markup Language (HTML), Cascades Style Sheets (CSS) and SMARTY J for balancing U3 shirt assembly line. A recent study, comprehensively evaluated 2 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx the garment assembly using Anylogic simulation software and has proved that there is need to do optimization because simulation technique is only descriptive and does not help in decision making [29]. Optimization method for assembly line balancing also has great limitations. This is normally common for manual or semi-automatic system like most apparel industries, as it is impossible to gain certain results with metaheuristics algorithms. Chen et al. [20] for example, conducted a study on grouping genetic algorithm to solve assembly line balancing problem with different labor skill levels in sewing lines of garment industry. Dinh et al.[30] applied greedy strategy to find an initial solution, followed by Simulated Annealing (SA) to find the best solutions for the garment assembly line balancing problem. While Xu et al.[31] applied an adaptive ant colony (AAC) algorithm with modifications made on the traditional ant colony algorithm for solving the Assembly line balancing problem. In this respect, to again the dual advantages and overcome the limitation of using either simulation or optimization techniques, simulation-based optimization is central. The distinctive hurdles for garment manufacturers are long production lead time, bottlenecking, and low productivity. Moreover, sewing or assembly line is the most critical phenomenon of garment manufacturing as it generally composes of huge number of operations [32]. The new paradigm shift in mass customization, the traditional garment production model needs to be optimized to have a more sustainable structure in order to meet demand for flexibility, low- cost, and high-efficiency [33]. The present paper is based on the empirical study conducted at Southern Range Nyanza Limited (NYTIL). The NYTIL garment facility regularly receives customers’ orders in large quantity which has imposed constant pressure to meet customers’ demand. However, the current operating line efficiency of 61.2% is too low for the company to achieve its goal even if the operators undergo forced-overtime. In order to address this problem, an exceptional study was conducted to balance garment assembly line using simulation-based optimization via design of experiment. The implicit mathematical formulation of the optimization problem is illustrated by Eqn. 1. 𝑀𝑎𝑥(𝑇ℎ𝑟𝑜𝑢𝑔ℎ𝑝𝑢𝑡) = 𝑀𝑎𝑥 {𝑓𝑖 (𝐴𝐶 , 𝐴𝑆 , 𝐴𝐿 )𝐴𝑆 , 𝐴𝐿 ∶ ℎ𝑖 (𝐴𝐶 , 𝐴𝑆 , 𝐴𝐿 )}𝐴𝑖 at constant ( 𝐴𝑆 , 𝐴𝐿 ) (1) where; 𝑓𝑖 (𝐴𝐶 , 𝐴𝑆 , 𝐴𝐿 ) = model input function, ℎ𝑖 (𝐴𝐶 , 𝐴𝑆 , 𝐴𝐿 ) = function of constraints on the model control factors (𝐴𝐶 ), model stochastic factors (𝐴𝑆 ) and model logic control (𝐴𝐿 ) and 𝐴𝑖 = set of model constraints. The current paper demonstrates assembly line balancing using simulation-based optimization via design of experiment without violating the set constraints and it is organized as follows; Section 2 briefly provide the relevant literature on the study, while section 3 describes the methodology of the study. Finally, section 4 presents the results and discussions of the study 2 | LITERATURE REVIEW Simulation-based optimization is also well-known as simulation optimization, black box optimization, parametric optimization, stochastic optimization, and optimization via simulation [34]. It is a state-of-art assembly line design approach that generate a number of scenarios from a probabilistic model and then select the best alternative solution by applying scheduling decisions and aggressive search approaches to these scenarios to obtain the best solution [35]. Simply put, simulation-based optimization is a combination or integration of simulation and optimization/metaheuristic techniques. The inherent complexity of assembly line and large number of feasible design alternatives make it extremely difficult to identify a global best solution with only simulation technique [37, 38]. For this reason, integrating simulation with optimization means that all the advantages 3 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx of the two design techniques can be harnessed. Moreover, simulation-based optimization can be applied to solve system design as well as the scheduling problem [38]. In general, it has been used to solve a number of industrial engineering problems [39]. Vast researchers have improved the performance of a number of systems using simulation- based optimization. Sarhangian et al. [40] for example, applied simulation-based metaheuristic optimization technique to optimize inspection strategies for multi-stage manufacturing processes. Dang and Phan [8] designed assembly line for footwear production using simulation-based optimization on Arena software. Juan [41] conducted study on production planning in manufacturing industry using simulation-based optimization. A study by Yegul et al. [42] improved the configuration of the production line using simulation-based optimization. Alvandi et al. [43] proposed an integrated simulation-optimization framework based on metaheuristic method to overcome an inherited complexity of classical production planning in multi-product/multi-machine production systems and optimizes several production objectives simultaneously. To some extend simulation-based optimization has been applied beyond manufacturing sector and thus covering transport, agriculture, defense and healthcare. For instance, Ibrahim et al. [44] conducted a study on minimization of patient waiting time in emergency department of a public hospital using simulation-based optimization approach. While, Shakibayifar [38] applied simulation-based optimization technique to rescheduling train traffic in uncertain conditions during disruptions. Likewise, Osorio et al.[45] applied Simulation-based optimization in transportation. Masoud et al.[46] demonstrated the applicability of simulation-based optimization in agriculture (horticultural nurseries). Ky et al.[47] reviewed surrogate-based method for black box optimization. While, Jeong et al.[48] integrated both metamodel and metaheuristics method for exploring design parameters in a defense system (hybrid system). So far so good, vast methods and tools or techniques for simulation-based optimization have been developed or used by the previous studies. Leung & Lau [23] for example, applied a multi- objective simulation-based optimization framework consisting of a hybrid immune-inspired algorithm named Suppression-controlled Multi-objective Immune Algorithm (SCMIA) and a simulation model for solving a real-life multi-objective optimization problem. Fatih et al. [49] optimized production line configuration, and proposed several simulation-based optimization approaches based on myopic search, ant-colony, simulated annealing and response-surface methodologies. González-reséndiz et al. [50] applied simulation and Response Surface Methodology (RSM) in optimizing logistics process for electronic goods. Jerbi et al.[51] revealed that Arena/OptQuest optimization platform outperforms the Taguchi optimization method. Chiadamrong & Piyathanavong [52] used OptQuest and Arena to search for the optimal supply chain network decisions under 3 levels of uncertainty. Similarly, Elnaggar [53] used Arena and OptQuest to determine the best number of workstations in garment assembly line. While Borodin et al.[54] demonstrated the possibility of integrating Arena and CPLEX software tools for simulation-based optimization. Evidence from the previous studies shows an integral of Arena and OptQuest as the most commonly used simulation-based optimization software tool [55]. However, in any design problem selection of software to be used for simulation study is very important and majorly based on the a number of criteria including the ease of use, animation capability, model development and input category [56]. 3 | METHODOLOGY 3.1 | STUDY APPROACH 4 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx The present study adopted a multidisciplinary research approach which comprised of both qualitative and quantitative research methods. It was conducted systematically following the several steps as illustrated in Figure 1. Figure 1 Methodology approach for the study 3.2 | CURRENT STATE ANALYSIS The current state analysis of NYTIL garment facility involved the system definition and conceptual model development. The system definition of trouser assembly line (Figure 2) was mainly done through observation and brainstorming of four categories of people: operators, quality personnel, maintenance personnel and line supervisors. The conceptual model of trouser assembly line was developed using process mapping method. The conceptual model (Figure 3) is simply a series of logical relationships relative to the components (a-q trouser parts) and structure of the trouser assembly line. This involved mapping all the processes or tasks associated with making trouser (1-72 tasks or operations). The assembly line was divided into eleven (11) sections including front, back preparation and main body assembling. 5 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx Figure 2 The NYTIL trouser assembly line 6 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx Figure 3 Conceptual model of Trouser assembly line 3.3 | MODELING OF INPUTS The trouser assembly line system consisted of 72 operations. However, each task is completed and repeated at different time. Therefore, to understand the operations of trouser assembly line, continuous stopwatch time study combined with observation was conducted [57]. The observed time in seconds was first converted to standard time units (minutes) and then recorded. The total of 60 observed times for each task were obtained comprising of 20 measurements recorded at interval of three periods of production season. Hence, a lot of variabilities in the tasks processing time were capture for Arena input modeling. The Arena input Analyzer was used to fit the distribution of these processing times for each operation involved in trouser assembly line. For instance, left front rise overlock operation has its fitted processing time distribution as illustrated in Figure 4. Similarly, all the other operations’ times were analyzed and summarized as shown in Table 1. Figure 4 Left front rise overlock fitted processing time distribution Table 1 Fitted processing time distribution OPN Operations description Resource Qty Processing time distribution per Bundle size resource 1 Left flybox pressing Iron press 1 TRIA (3, 5.12, 5.9) 25 2 Buttonhole on Left flybox BH 1 6.05 + ERLA (0.39, 6) 25 3 Left front rise overlock 3t O/L 1 4 + 6.88 * BETA (1.95, 3.37) 25 4 Right front rise overlocks 2.29 + ERLA (0.239, 5) 25 5 Knee patch attach S/NL 3 20 + 21 * BETA (0.856, 1.33) 25 6 Side pocket flatlock F/L 2 4 + 4 * BETA (1.94, 2.74) 25 7 Side pocket overlocks 5t O/L 1 2 + ERLA (0.555, 2) 25 8 Right flybox overlock 1.6 + LOGN (0.719, 0.418) 9 Side pocket attach S/NL 2 7 + 11 * BETA (1.67, 1.67) 25 10 Side pocket topstitch S/NL 2 10 + GAMM (1.44, 2.7) 25 11 Right flybox attach S/NL 2 TRIA (13, 20.7, 25) 25 12 Left fly box tacking S/NL 2 9 + WEIB (3.39, 2.09) 25 13 Fly attach S/NL 2 12.1 + GAMM (0.955, 3.94) 25 14 Front prep bundling Helper 1 5 + 10 * BETA (1.27, 2.07) 25 15 Back marking Helper 1 3 + 4.65 * BETA (1.55, 2.76) 25 16 Back patch pressing Iron press 1 TRIA (3, 8.29, 9.73) 25 17 Back patch attach S/NL 2 10 + 11 * BETA (0.737, 0.96) 25 18 Hip pocket cutting AWM 1 TRIA (3.17, 3.99, 7) 25 19 Hip pocket overlocks 5t O/L 1 5 + 3.83 * BETA (2.14, 3.14) 25 20 Hip flap folding Helper 1 NORM (4.77, 0.65) 25 21 Button Hole on hip flap BH 1 3.55 + GAMM (0.194, 5.47) 25 22 Hip flap runstitch S/NL 1 3 + LOGN (2.72, 1.83) 25 23 Hip flap turning TM 1 NORM (3.25, 0.551) 25 24 Hip flap topstitches S/NL 1 3 + 5 * BETA (1.7, 1.88) 25 25 Hip flap attach S/NL 2 5.45 + LOGN (1.44, 0.936) 25 26 Hip pocket finish 19 + 10 * BETA (1.46, 1.46) 27 Back prep bundling Helper 1 3 + 2 * BETA (0.889, 0.968) 25 7 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx OPN Operations description Resource Qty Processing time distribution per Bundle size resource 28 Front and back bundling Helper 1 2 + 6.86 * BETA (1.18, 2.11) 25 29 Side seam overlock 5t O/L 2 NORM (1.21, 0.115) Not bundled 30 Side seam topstitch F/A 2 TRIA (0.52, 0.747, 0.94) Not bundled 31 Knee pocket point marking Helper 1 0.32 + 0.57 * BETA (0.889, 1.18) Not bundled 32 Knee pocket topstitch S/NL 2 11 + ERLA (1.89, 2) 25 33 Knee pocket tacking S/NL 1 4 + 3 * BETA (1.33, 1.75) 25 34 Knee pocket Overlock 5t O/L 1 2 + 4 * BETA (0.831, 2.05) 25 35 Knee pocket hemming S/NL 1 2 + 4 * BETA (1.41, 1.13) 25 36 Knee pocket ironing Iron press 2 8 + 5.78 * BETA (0.957, 1.06) 25 37 Knee pocket attach S/NL 2 0.88 + 0.92 * BETA (1.77, 1.96) Not bundled 38 Knee flap folding Helper 1 3.63 + 3.13 * BETA (3.89, 2.38) 25 39 Button hole on knee flap BH 1 4.27 + WEIB (1.21, 1.99) 25 40 Knee flap runstitch S/N 1 TRIA (2.37, 3.81, 6.88) 25 41 Knee flap turning TM 1 NORM (4.02, 1.01) 25 42 Knee flap topstitch S/NL 1 4 + 5.78 * BETA (0.903, 2.11) 25 43 Knee flap attach D/NL 2 TRIA (0.67, 1.04, 1.7) Not bundled 44 Bar tacking BT 2 NORM (1.25, 0.266) Not bundled 45 Back rise overlocks 5t O/L 1 0.26 + LOGN (0.185, 0.0881) Not bundled 46 Back rise Topstitch D/NL 1 NORM (0.439, 0.0494) Not bundled 47 Big loop matching Helper 1 NORM (0.0663, 0.018) Not bundled 48 Big loop runstitch S/NL 3 0.12 + 0.3 * BETA (2.89, 5.28) Not bundled 49 Big loop turning Helper 2 0.07 + GAMM (0.0143, 7.47) Not bundled 50 Big loop runstitch S/NL 2 0.09 + 0.19 * BETA (1.78, 2) Not bundled 51 Big loop button hole BH 1 TRIA (0.04, 0.055, 0.11) Not bundled 52 Small loop runstitch LM 1 TRIA (0.11, 0.134, 0.18) Not bundled 53 Small loop, big loop and S/NL 3 1.58 + ERLA (0.068, 7) Not bundled waistband attach 54 Waistband topstitch S/NL 2 TRIA (0.73, 1.34, 1.5) Not bundled 55 Waist band closing with size and S/NL 2 0.77 + GAMM (0.0607, 3.58) Not bundled label tags 56 Inseam Overlock 5t O/L 2 0.49 + WEIB (0.483, 6.16) Not bundled 57 Trouser turning Helper 1 0.2 + LOGN (0.218, 0.112) Not bundled 58 Inseam topstitch F/A 2 0.32 + 0.56 * BETA (1.98, 1.61) Not Bundled 59 Button hole on Hip band BH 1 TRIA (0.31, 0.344, 0.47) Not bundled 60 Button hole on the bottom leg BH 1 0.32 + 0.2 * BETA (2.7, 3.33) Not bundled 61 Bottom rope attach Helper 1 0.5 + LOGN (0.251, 0.168) Not bundled 62 Bottom hemming S/NL 2 0.71 + 0.73 * BETA (2.04, 2.6) Not bundled 63 Small loop tacking S/NL 2 TRIA (0.82, 1.17, 1.37) Not bundled 64 Final Bar tacking BT 2 TRIA (0.74, 0.851, 1.05) Not bundled 65 Adjustable rope cutting Helper 1 TRIA (0.1, 0.145, 0.19) Not bundled 66 Adjustable hemming S/NL 1 TRIA (0.1, 0.136, 0.2) Not bundled 67 1st adjustable rope attach S/NL 1 NORM (0.75, 0.0479) Not bundled 68 2nd adjustable rope attach S/NL 1 0.53 + 0.32 * BETA (3.19, 2.1) Not bundled 69 Button point marking Helper 1 0.55 + GAMM (0.0328, 6.16) Not bundled 70 Trimming Helper 7 NORM (4.84, 0.345) Not bundled 71 Quality checking QP 2 0.82 + LOGN (0.332, 0.154) Not bundled 72 Rework S/NL 1 TRIA (2, 3.5, 4.7) Not bundled OPN- Operation Number, S/NL- Single needle lockstitch, BH- Button hole machine, F/A- Feed of arm, 5t O/L- 5 threads overlock machine, LM- loop stitching machine, D/NL- Double needle lockstitch, BT- Bartack machine, QP- quality personnel, TM- turning machine, AWM- automatic wallet machine, 3t O/L- 3 thread overlock machine, F/L- Flatlock machine, Qty- Quantity 3.4 | CONSTRUCTION OF COMPUTER MODEL The computer model of the trouser assembly line was developed based on the discrete event simulation using Arena simulation software (Academic license version 16). The simulation model was built on a 64 bits notebook computer with a 2.00 GHz Intel core i3 CPU and 4.00 GB RAM. Due to the low processing speed of the notebook computer, 32 bits Arena software category was well-suited to be installed instead of the 64 bits. Consequently, the simulation model was developed and run smoothly without freezing the computer. For the computer model 8 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx construction, the following model input were essential: the processing times, number of machines, number of operators, number of tasks, number of helpers, quantity of input material per day, interarrival time of parts, productivity per day, working hours, quantity of input materials per day, task precedence relations, bundle sizes, job release policy, machine type and production target. In the construction of computer model or existing/base model (Figure 5) [58], several Arena simulation elements were used including entity, variables, resources, process, attribute and transfer and control logic elements. The following model assumptions were used for simulation model development in this present study: (i) The input materials arrive in production line at constant time i.e. every day and there was no shortage of material from the cutting section. (ii) There was no breakdown of the machines in the production line (iii) There are no absenteeism of the operators and so machines are never stopped due to absence of the operators. (iv) Each operator and helper were assigned to perform a single task in the production line and only operators were assigned to sewing machines. Therefore, the number of operators were equal to the number of machine and increasing machine number also increases operator number and vice versa. (v) Production only runs for 8 hours in a day and there was no overtime. (vi) All defected trousers at 8% defect rate per day were reworked by only one workstation with a single needle lockstitch machine. Arena modules used for discrete event simulation development Figure 5 A section of Arena discrete event simulation for trouser assembly line 9 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx 3.5 | VERIFICATION AND VALIDATION After developing the computer model, the simulation runs were conducted to verify if the model follows the logic pointed out in the conceptual model. Verification is basically the process of ensuring that the model behaves as intended [59]. More specifically it is known as debugging the model. Therefore, Trace and animation techniques were used to verify that each program path is correct (Figure 6). Further, the model verification was done through testing and observing the simulation model at varying situation including changes made on the interarrival time, process time, run length and replication time. Figure 6 Arena animation of full trouser assembly line operations The simulation replication length (n=10) was determined in accordance to Kelton et al.[59] with the run length of one month (i.e. 28 days of 8 hours daily production). In this study, the steady-state simulation with 2 days warm-up period was approximated according to Law [60]. Then, the Arena simulation run was executed, and the throughput (𝜇𝐴 = 496 pieces per day) and half width (6.61) were obtained. The hypothesized mean (𝜇𝐴 ) was used for comparison with real-world system throughput samples. The real-world system throughput with sample size (N=23) collected for a period of one month from real-world system was used for validation of the operation model. One-sample-T hypothesis test at 95% confidence interval (CI) was successfully accomplished with the help of Minitab statistical software version 18. Since 𝜇𝐴 lies within 95% CI for real-world system average throughput (𝜇𝑅 ), the null hypothesis (𝐻0 ) was accepted with the T-value (-0.2) and P-value (0.842) as shown in Table 2. Hence, the 10 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx Arena simulation model of trouser assembly line was validated including all the assumptions used in building the model. Table 2 Descriptive statistic for real-world throughput sample and hypothesis test N Mean StDev SE Mean 95% CI for 𝝁𝑹 T-value P-value Null Alternative (𝝁𝑹 ) hypothesis hypothesis 23 490.8 124.2 25.9 (437.1, 544.5) -0.20 0.842 𝜇𝑅 = 𝜇𝐴 , 𝐻0 𝜇𝑅 ≠ 𝜇𝐴 , 𝐻1 N- number of throughput samples from real-world system, 4 | RESULTS AND DISCUSSIONS 4.1 | DESIGN SCENARIOS The resolution-V experimental design with five factors (bundle size, job release policy, task assignment pattern, machine number and helper number) and two levels was used to generate sixteen (16) assembly line design scenarios. For each scenario, was simulation experiments was executed and the response (throughput) was observed for each case as shown in Figure 7 [58]. Scenario 2 produced the highest average throughput and it was selected to be the local best (optimal) solution for trouser assembly line balancing. The scenario 2 was adopted as the starting solution for OptQuest optimization process. 700 638 609 607 583 600 496 496 496 496 Throughput per day 500 465 467 467 467 467 467 429 439 400 300 200 100 0 Figure 7. Assembly line balancing scenarios based on experimental design 4.2 | OPTQUEST OPTIMIZATION A black box optimization on the trouser assembly line simulation model was performed using OptQuest for Arena. The objective function for the optimization model was to maximize the throughput, and only two control factors (machine number and helper number) were considered. This is because the other three factors were determined to be statistically insignificant by analysis of variance (ANOVA) method. The optimization constraints included Machine number ≤ 10, Helper number ≤ 5 and Throughput ≤ 800. While the lower and upper bound was set to be 1and 3 respectively (i.e. 1≤ x ≤3) for each sewing machine type and helper at specific workstation. To this end, the optimization model elements: objective function, controls, constraints, lower and upper bounds were fed into the OptQuest. An Optimization process was executed with 11 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx automatic termination of approximately 100 simulations. Each simulation presented further different design scenarios with average throughput (best values) as illustrated in Figure 8. Figure 8 OptQuest optimization results At the end of the Optimization process, OptQuest presented 20 best solutions for each simulation out of 100 total simulations performed as presented in Table 3. The differences in the objective value (throughput) among these best solutions are negligible because they differ infinitesimally. Therefore, it was a great challenge to select the global optimal solution from the 20 candidate’s best solutions. Nevertheless, three outstanding best solutions were selected by consideration of resource numbers. In this respect, they utilized fewer resources than other simulations. However, from the three selected candidate’s solutions (Simulation 33, 32 and 8), further decision was taken to obtain global best solution. In order to unlock this, separate Arena simulation model was created for each case and the simulation experiments were performed to observe the work in progress (WIP) at workstations. Simulation 33 is show balanced assembly line but it utilizes more resource than simulation 32 and simulation 8, therefore, it was eliminated. While simulation 8 has less resource than simulation, it accumulated high WIP at bartacking workstation. For this reason, simulation 32 was determined to be the global optimal solution because its design resulted into well-balanced assembly line with low WIP. Table 3 Best solutions from OptQuest optimization Number of resources (machine and helper) added Simulation Objective 5threads Bartack Helper Single needle lockstitch value overlock machine 80 762.3 1 2 2 4 40 762.2 1 1 2 5 43 762.2 1 1 2 5 33 762.1 1 1 2 4 81 762.1 2 2 2 3 45 762 1 1 2 4 20 761.9 1 0 2 5 32 761.8 1 1 2 3 74 761.6 1 2 2 4 77 761.6 1 2 2 3 12 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx Number of resources (machine and helper) added Simulation Objective 5threads Bartack Helper Single needle lockstitch value overlock machine 18 761.4 2 1 2 4 29 761.3 1 1 2 5 79 761.3 1 0 2 6 28 761.2 2 0 2 4 42 761.1 2 0 2 5 34 761 1 0 2 5 41 761 0 1 2 6 8 760.9 1 0 2 3 30 760.9 1 0 2 5 71 760.7 1 2 2 5 The operator numbers are assumed to be equal to the machine numbers for all line balancing cases because one operator was allowed to operate one machine on trouser assembly line. Since the assumption made was that increasing or decreasing machine number automatically changes the operator number, vice versa. In the optimization process, the constant bundle size of 25 was used because the effect of varying its level (10 to 40 bundle sizes) was statistically insignificant. Further, it was noted that increasing resource number has direct effect on the average throughput of the assembly line. The average throughput of 638 pieces per day was achieved at local optimal solution and 762 at global optimal with the increase of line efficiency to 79.7% and 95.2% respectively. A reflection from the previous studies, Anisah et al.[61] achieved an increase in the average throughput at local optimal solution when resources were added. Additionally, Yemane et al. [7] achieved line efficiency of 75.3% at the local optimal solution. For both of these studies, optimization process was not conducted. Nevertheless, it is an evident that increase of resource number leads to increase of throughput and line efficiency. However, it is only true when resources are added at the bottleneck workstations. Notably, the bottleneck workstations are effectively determined through extensive simulation of the assembly line. In order for NYTIL company to achieve the increase in the production throughput and efficiency, an optimal design is required to be implemented in their trouser assembly line as shown in Table 4. Please see Appendix, it presents more details on the resource allocation per workstations for the existing, local and global optimal design. Table 4 Comparison of the three trouser assembly line design stages S/N Resource type Resource number Initial/existing Local optimal Global optimal 1. Single needle lockstitch machine 47 50 53 2. Double needle lockstitch machine 3 3 3 3. Flatlock machine 1 1 1 4. Overlock machine (3 and 5threads) 9 9 10 5. Feed of arm 4 4 4 6. Automatic wallet machine 1 1 1 7. Iron press machine 3 4 4 8. Bartack machine 4 4 5 13 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx S/N Resource type Resource number Initial/existing Local optimal Global optimal 9. Button hole sewing machine 6 6 6 10. Small loop sewing machine 1 1 1 11. Turning machine 4 4 4 12. Operator 83 87 92 13. Helper 19 22 24 14. Quality personnel 2 2 2 Total 187 198 210 CONCLUSION The present study has successfully demonstrated garment assembly line design using simulation-based optimization via design of experiment. The conceptual model was constructed based on the current practice in trouser assembly line which was validated by line supervisors. Most details on trouser production process was captured during conceptual model construction which simplified the development of the simulation model. The study showed that simulation model is an acceptable approximate of real-world trouser assembly line at 95% confidence interval. Increasing resource (machine and helper) numbers at the bottleneck workstations increased the average production throughput and efficiency at both local and global optimal assembly design. The high increase in line efficiency and average throughput confirmed the suitability assembly line balancing using simulation-based optimization via design of experiment. This is because the local optimal solution obtained at simulation experimentation phase narrowed down the search space of the OptQuest optimization process, hence, OptQuest rapidly and aggressively reached the global optimal solution. However, nonlinear single objective optimization with discrete control values was considered in the present study which imposed conflicting decisions on the global optimal solution among the candidates’ best solutions. Therefore, further study should develop a profound optimization model (multi-objective) with at least two objective functions including production cost, cycle time and resource utilization. ACKNOWLEDGEMENTS The authors acknowledge the financial support from the ACE II-PTRE of Moi University (Credit No. 5798-KE), Kenya, toward the research project. Furthermore, the authors acknowledge the Rockwell Automation for providing Arena software academic research license version 16. The authors are also grateful to the management of Southern range nyanza limited (NYTIL) for allowing the research to be conducted in their company. CONFLICT OF INTEREST The authors declare no potential conflict of interest. REFERENCES [1] W. Grzechca and L. R. Foulds, “The Assembly Line Balancing Problem with Task Splitting : A Case Study,” IFAC-PapersOnLine, vol. 48, no. 3, pp. 2002–2008, 2015. DOI: https://doi.org/10.1016/j.ifacol.2015.06.382. 14 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx [2] N. T. Thomopoulos, Assembly Line Planning and Control. Springer Cham Heidelberg New York Dordrecht London: Springer International Publishing Switzerland, 2014. DOI: https://doi.org/10.1007/978-3-319-01399-2. [3] A. Dolgui and J. Proth, “Assembly Line Balancing : Conventional Methods and Extensions,” in 7th IFAC Conference on Manufacturing Modelling, Management, and Control, 2013, vol. 46, no. 9, pp. 43–48. DOI: https://doi.org/10.3182/20130619-3-RU- 3018.00644. [4] A. A. Alghazi, “Balancing and sequencing of assembly lines,” Published D.Phil Thesis at All Dissertations 2022. Clemson University, 2017. DOI: https://doi.org/10.1007/978- 3-662-00861-4. [5] M. Azizo and M. Z. Min, “Research Workload smoothing in simple assembly line balancing,” Comput. Oper. Res., vol. 89, pp. 51–57, 2018. DOI: https://doi.org/10.1016/j.cor.2017.08.006. [6] A. Dolgui and E. Gafarov, “Some new ideas for assembly line balancing research,” IFAC-PapersOnLine, vol. 50, no. 1, pp. 2255–2259, 2017. DOI: https://doi.org/10.1016/j.ifacol.2017.08.189. [7] A. Yemane, S. Haque, and I. S. Malfanti, “Optimal Layout Design by Line Balancing Using Simulation Modeling,” in Proceedings of the International Conference on Industrial Engineering and Operations Management, 2017, pp. 228-245. [8] Q. V. Dang and K. Pham, “Design of a Footwear Assembly Line Using Simulation- based ALNS,” in 13th Global Conference on Sustainable Manufacturing - Decoupling Growth from Resource Use, 2016, pp. 596–601. DOI: https://doi.org/10.1016/j.procir.2016.01.140. [9] X. Delorme, A. Dolgui, S. Kovalev, and M. Y. Kovalyov, “Minimizing the Number of Workers in a Paced Mixed-Model Assembly Line,” Eur. J. Oper. Res., 2018. DOI: https://doi.org/10.1016/j.ejor.2018.05.072. [10] E. Alvarez-miranda and J. Pereira, “On the complexity of assembly line balancing problems,” Comput. Oper. Res., 2019. DOI: https://doi.org/10.1016/j.cor.2019.04.005. [11] Y. Li, Y. Fu, X. Tang, and X. Hu, “Optimizing the Reliability and Efficiency for an Assembly Line That Considers Uncertain Task Time Attributes,” IEEE Access, vol. 7, pp. 34121–34130, 2019. DOI: https://doi.org/10.1109/ACCESS.2019.2897730. [12] M. Gansterer and R. F. Hartl, “One- and two-sided assembly line balancing problems with real-world constraints,” Int. J. Prod. Res., vol. 7543, pp. 1–18, 2018. DOI: https://doi.org/10.1080/00207543.2017.1394599. [13] M. Jusop, M. F. F. A. Rashid, and N. H. Kamarudin, “A simulation-based approach for solving assembly line balancing problem A simulation-based approach for solving assembly line balancing problem,” IOP Conf. Ser. Mater. Sci. Eng., vol. 231, no. 012189, pp. 1–6, 2017. DOI: https://doi.org/10.1088/1757-899X/231/1/012189. [14] X. Zhang and Y. Chen, “Carbon Emission Evaluation Based on Multi-Objective Balance of Sewing Assembly Line in Apparel Industry,” Energies, vol. 12, no. 2783, pp. 1–19, 2019. DOI: https://doi.org/10.3390/en12142783. [15] G. Karabay, “A Comparative Study on Designing of a Clothing Assembly Line,” Tekst. 15 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx Ve Konfeksiyon, vol. 24, no. 1, pp. 124–133, 2014. [16] O. Bongomin, J. I. Mwasiagi, E. O. Nganyi, and I. Nibikora, “Improvement of garment assembly line efficiency using line balancing technique,” Eng. Reports, vol. 2, no. 4, pp. 1–18, 2020. DOI: https://doi.org/10.1002/eng2.12157. [17] T. Aysun, Y. Yalcin, and K. Mahmut, “Heuristic production line balancing problem solution with MATLAB software programming,” Int. J. Cloth. Sci. Technol., pp. 1–31, 2016. DOI: https://doi.org/10.1108/IJCST-01-2016-0002. [18] P. Umarani and K. Valase, “Assembly line balancing in textile industry,” Int. J. Sci. Res. Eng. Technol., vol. 6, no. 4, pp. 323–329, 2017. [19] H. Simea, P. Janab, and D. Pangha, “Feasibility of Using Simulation Technique for Line Balancing in Apparel Industry,” Procedia Manuf., vol. 30, pp. 300–307, 2019. DOI: https://doi.org/10.1016/j.promfg.2019.02.043. [20] J. C. Chen et al., “Assembly Line Balancing Problem of Sewing Lines in Garment Industry,” in International Conference on Industrial Engineering and Operations Management, 2014, pp. 1215–1225. DOI: https://doi.org/10.1016/j.eswa.2012.02.055. [21] S. H. Eryuruk, “Clothign Assembly line Design Using Simulation and Heuristic line Balancing Techniques,” Tekst. Ve Konfeksiyon, vol. 4, pp. 360–368, 2012. [22] O. Bongomin, G. G. Ocen, E. O. Nganyi, A. Musinguzi, and T. Omara, “Exponential Disruptive Technologies and the Required Skills of Industry 4.0,” J. Eng., vol. 2020, pp. 1–17, 2020. DOI: https://doi.org/10.1155/2020/4280156. [23] C. S. K. Leung and H. Y. K. Lau, “Multiobjective Simulation-Based Optimization Based on Artificial Immune Systems for a Distribution Center,” J. Optim., vol. 2018, pp. 1–15, 2018. DOI: https://doi.org/10.1155/2018/5852469. [24] R. Liu, X. Xie, K. Yu, and Q. Hu, “A survey on simulation optimization for the manufacturing system operation,” Int. J. Model. Simul., vol. 6203, pp. 1–12, 2018. DOI: https://doi.org/10.1080/02286203.2017.1401418. [25] D. Kitaw, A. Matebu, and S. Tadesse, “Assembly Line Balancing Using Simulation Technique in a Garment Manufacturing Firm,” J. EEA, vol. 27, pp. 69–80, 2010. [26] S. Kursun and F. Kalaoglu, “Simulation of production line balancing in apparel manufacturing,” Fibres Text. East. Eur., vol. 17, no. 4(75), pp. 68–71, 2009. [27] P. Stief, J. Dantan, A. Etienne, and A. Siadat, “Simulation-based Planning of a Kind of Complex Product General Assembly Line,” Procedia CIRP, vol. 76, pp. 25–30, 2018. DOI: https://doi.org/10.1016/j.procir.2018.01.032. [28] O. A. Hanan and A. I. Seedahmed, “Garment Assembly Line Balancing Using Modeling and Simulation,” Int. J. Sci. Eng. Sci., vol. 3, no. 3, pp. 4–7, 2019. [29] Y. Xu, S. Thomassey, and X. Zeng, “Comprehensive evaluation of garment assembly line with simulation,” in IOP Conference Series: Materials science and engineering, 2017, pp. 1–6. DOI: https://doi.org/10.1088/1757-899X/254/16/162013. [30] M. H. Dinh, V. L. Truong, T. T. Phan, V. D. Nguyen, P. T. Do, and D. N. Nguyen, “Simulated Annealing for the Assembly Line Balancing Problem in the Garment Industry,” in The Tenth International Symposium on Information and Communication 16 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx Technology (SoICT 2019), December 4–6, 2019, 2019, pp. 36–42. DOI: https://doi.org/10.1145/3368926.3369698. [31] H. Xu, B. Xu, and J. Yan, “Balancing apparel assembly lines through adaptive ant colony optimization,” Text. Res. J., vol. 0, no. 00, pp. 1–15, 2018. DOI: https://doi.org/10.1177/0040517518819836. [32] A. Yemane, G. Gebremicheal, T. Meraha, and M. Hailemicheal, “Productivity Improvement through Line Balancing by Using Simulation Modeling ( Case study Almeda Garment Factory ),” J. Optim. Ind. Eng., vol. 13, no. 1, pp. 153–165, 2020. DOI: https://doi.org/10.22094/JOIE.2019.567816.1565. [33] K. Ma, L. Wang, and Y. Chen, “A Resource Sharing Mechanism for Sustainable Production in the Garment Industry,” Sustainability, vol. 10, no. 52, pp. 1–22, 2018. DOI: https://doi.org/10.3390/su10010052. [34] S. Amaran, N. V Sahinidis, B. Sharda, and S. J. Bury, “Simulation optimization : A review of algorithms and applications,” pp. 1–28, 2017. DOI: https://doi.org/10.1007/s10479-015-2019-x. [35] M. Dehghanimohammadabadi, T. K. Keyser, and S. H. Cheraghi, “A novel Iterative Optimization-based Simulation (IOS) framework: An effective tool to optimize system’s performance,” Comput. Ind. Eng., pp. 1–17, 2017. DOI: https://doi.org/10.1016/j.cie.2017.06.037. [36] G. Michalos, A. Fysikopoulos, S. Makris, D. Mourtzis, and G. Chryssolouris, “Multi criteria assembly line design and configuration - An automotive case study,” CIRP J. Manuf. Sci. Technol., vol. 9, pp. 69–87, 2015. DOI: https://doi.org/10.1016/j.cirpj.2015.01.002. [37] A. T. Bon and N. N. Shahrin, “Assembly Line Optimization using Arena Simulation,” in Proceedings of the 2016 International Conference on Industrial Engineering and Operations Management, 2016, pp. 2225–2232. [38] M. Shakibayifar, A. Sheikholeslami, and F. Corman, “A simulation-based optimization approach to rescheduling train tra c in uncertain conditions during disruptions,” Sci. Iran., vol. 25, no. 2, pp. 646–662, 2018. DOI: https://doi.org/10.24200/sci.2017.4186. [39] W. T. D. S. Junior, J. A. B. Montevechia, R. de C. Miranda, and A. T. Campos, “Discrete simulation-based optimization methods for industrial engineering problems : A systematic literature review,” Comput. Ind. Eng., vol. 128, no. December 2018, pp. 526– 540, 2019. DOI: https://doi.org/10.1016/j.cie.2018.12.073. [40] V. Sarhangian, A. Vaghefi, H. Eskandari, and M. K. Ardakani, “Optimizing inspection strategies for multi-stage manufacturing processes using simulation optimization,” in Proceedings of the 2008 Winter Simulation Conference, 2008, pp. 1974–1980. DOI: https://doi.org/10.1109/WSC.2008.4736291. [41] E. L. D. Juan, “Simulation-Based Optimization for Production Planning: Integrating Meta-Heuristics, Simulation and Exact Techniques to Address the Uncertainty and Complexity of Manufacturing Systems,” D.Phil. Thesis: Alliance Manchester Business School, 2016. [42] M. F. Yegul, F. S. Erenay, S. Striepe, and M. Yavuz, “Improving configuration of complex production lines via simulation-based optimization,” Comput. Ind. Eng., vol. 17 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx 109, pp. 295–312, 2017. DOI: https://doi.org/10.1016/j.cie.2017.04.019. [43] S. Alvandi, W. Li, and S. Kara, “An Integrated Simulation Optimisation Decision Support Tool for Multi-Product Production Systems,” Mod. Appl. Sci., vol. 11, no. 6, pp. 56–71, 2017. DOI: https://doi.org/10.5539/mas.v11n6p56. [44] I. M. Ibrahim, C.-Y. Liong, S. A. Bakar, N. Ahmad, and A. F. Najmuddin, “Minimizing patient waiting time in emergency department of public hospital using simulation optimization approach,” in The 4th International Conference on Mathematical Sciences, 2017, vol. 1830, pp. 1-8. DOI: https://doi.org/10.1063/1.4980949. [45] C. Osorio, G. Flötteröd, and C. Zhang, “A metamodel simulation-based optimization approach for the efficient calibration of stochastic traffic simulators,” Transp. Res. Procedia, vol. 6, pp. 213–223, 2015. DOI: https://doi.org/10.1016/j.trpro.2015.03.017. [46] S. Masoud, Y.-J. Son, C. Kubota, and R. Tronstad, “Evaluation of Simulation-Based Optimization in Grafting Labor Allocation,” Appl. Eng. Agric., vol. 34, no. 3, pp. 479– 489, 2018. DOI: https://doi.org/10.13031/aea.12487. [47] V. K. Ky, C. D. Ambrosio, Y. Hamadi, and L. Liberti, “Surrogate-based methods for black-box optimization,” Int. Trans. Oper. Res., pp. 1–26, 2016. [48] H. H. Jeong, S. Kyung-Min, and T. G. Kim, “Simulation-based optimization for design parameter exploration in hybrid system: a defense system example,” Simul. Trans. of the Soc. Model. Simul. Int., pp. 1–19, 2013. DOI: https://doi.org/10.1177/0037549712466707. [49] M. Fatih, F. Safa, S. Striepe, and M. Yavuz, “Computers & Industrial Engineering Improving configuration of complex production lines via simulation-based optimization,” Comput. Ind. Eng., vol. 109, pp. 295–297, 2017. DOI: https://doi.org/10.1016/j.cie.2017.04.019. [50] J. González-reséndiz, K. C. Arredondo-soto, A. Realyvásquez-vargas, H. Híjar-rivera, and T. Carrillo-gutiérrez, “Integrating Simulation-Based Optimization for Lean Logistics : A Case Study,” Appl. Sci., vol. 8, no. 2448, pp. 1–18, 2018. DOI: https://doi.org/10.3390/app8122448. [51] A. Jerbi, A. Ammar, and M. Krid, “Performance optimization of a flexible manufacturing system using simulation : the Taguchi method versus OptQuest,” Simul. Trans. of the Soc. Model. Simul. Int., vol. 00, no. 0, pp. 1–12, 2019. DOI: https://doi.org/10.1177/0037549718819804. [52] N. Chiadamrong and V. Piyathanavong, “Optimal design of supply chain network under uncertainty environment using hybrid analytical and simulation modeling approach,” J. Ind. Eng. Int., vol. 13, no. 4, pp. 465–478, 2017. DOI: https://doi.org/10.1007/s40092- 017-0201-2. [53] G. Elnaggar, “Effect of Operator Skill Level on Assembly Line Balancing in Apparel Manufacturing : A Multi-Objective Simulation Optimization Approach,” in Proceedings of the International Conference on Industrial Engineering and Operations Management, 2019, pp. 308–315. [54] V. Borodin, J. Bourtembourg, F. Hnaien, and N. Labadie, “COTS software integration for simulation optimization coupling : case of ARENA and CPLEX products,” Int. J. Model. Simul., vol. 39, no. 3, pp. 178–189, 2019. DOI: 18 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx https://doi.org/10.1080/02286203.2018.1547814. [55] N. Prajapat and A. Tiwari, “A review of assembly optimisation applications using discrete event simulation,” Int. J. Comput. Integr. Manuf., vol. 30, no. 2–3, pp. 215–228, 2017. DOI: https://doi.org/10.1080/0951192X.2016.1145812. [56] A. Magno, C. Guimarães, J. E. Leal, and P. Mendes, “Discrete-event simulation software selection for manufacturing based on the maturity model,” Comput. Ind., vol. 103, pp. 14–27, 2018. DOI: https://doi.org/10.1016/j.compind.2018.09.005. [57] A. P. Puvanasvaran, C. Z. Mei, and V. A. Alagendran, “Overall Equipment Efficiency Improvement Using Time Study in an Aerospace Industry,” Procedia Eng., vol. 68, pp. 271–277, 2013. DOI: https://doi.org/10.1016/j.proeng.2013.12.179. [58] O. Bongomin, J. I. Mwasiagi, E. O. Nganyi, and I. Nibikora, “Optimal design of garment assembly line using simulation-based optimization, Mendeley data V1.” Mendeley dataset, 19-Oct-2019. DOI. https://doi.org/10.17632/T5W96KH5W7.1. [59] W. D. Kelton, R. P. Sadowski, and D. T. Sturrock, Simulation with Arena, 4th ed. New York: McGraw-Hill, 2007. [60] A. M. Law, “Statistical Analysis of simulation output data: The practical state of the Art,” in Proceedings of the 2007 Winter Simulation Conference S., 2007, pp. 77–83. [61] S. Anisah, A. Yaakub, R. Ramlan, and T. G. Foong, “Cycle Time Reduction of a Garment Manufacturing Company Using Simulation Technique,” in Proceedings International Conference of Technology Management, Business and Entrepreneurship 2012, 2012, no. 1–9. APPENDIX Details comparison of existing, local optimal and global optimal designs based on resource number allocation in the workstations WSN OPS Operations description Resource Type Number Existing Local Global design optimal Optimal 1 1 Left flybox pressing Iron press 1shared 1 1 2 2 Buttonhole on Left flybox Buttonhole machine 1 1 1 3 3 Left front rise overlock 3 threads overlock 1 1 1 4 Right front rise overlocks 4 5 Knee patch attach Single needle lockstitch 3 4 4 5 6 Side pocket flatlock Flatlock machine 1 1 1 6 7 Side pocket overlocks 5 threads overlock 1 1 1 8 Right flybox overlock 7 9 Side pocket attach Single needle lockstitch 2 2 2 8 10 Side pocket topstitch Single needle lockstitch 2 2 2 9 11 Right flybox attach Single needle lockstitch 2 2 2 10 12 Left fly box tacking Single needle lockstitch 2 2 2 11 13 Fly attach Single needle lockstitch 2 2 2 12 14 Front prep bundling Helper 1 1 1 13 15 Back marking Helper 1 1 1 14 16 Back patch pressing Iron press 1shared 1 1 15 17 Back patch attach Single needle lockstitch 2 2 2 16 18 Hip pocket cutting Automatic wallet machine 1 1 1 17 19 Hip pocket overlocks 5 threads overlock 1 1 1 18 20 Hip flap folding Helper 1 1 1 19 21 Button Hole on hip flap Button hole machine 1 1 1 19 Preprint Received: Added at production Revised: Added at production Accepted: Added at production DOI: xxx/xxxx 20 22 Hip flap runstitch Single needle lockstitch 1 1 1 21 23 Hip flap turning Turning machine 1 1 1 22 24 Hip flap topstitches Single needle lockstitch 1 1 1 23 25 Hip flap attach & Single needle lockstitch 2 3 3 26 Hip pocket finish 25 27 Back prep bundling Helper 1 1 1 26 28 Front and back bundling Helper 1 1 1 27 29 Side seam overlock 5 threads overlock 2 2 3 28 30 Side seam topstitch Feed of Arm 2 2 2 29 31 Knee pocket point marking Helper 1 1 2 30 32 Knee pocket topstitch Single needle lockstitch 2 2 2 31 33 Knee pocket tacking Single needle lockstitch 1 1 1 32 34 Knee pocket Overlock 5 threads overlock 1 1 1 33 35 Knee pocket hemming Single needle lockstitch 1 1 1 34 36 Knee pocket ironing Iron press 2 2 2 35 37 Knee pocket attach Single needle lockstitch 2 2 3 36 38 Knee flap folding Helper 1 1 1 37 39 Button hole on knee flap Button hole machine 1 1 1 38 40 Knee flap runstitch Single needle lockstitch 1 1 1 39 41 Knee flap turning Turning machine 1 1 1 40 42 Knee flap topstitch Single needle lockstitch 1 1 1 41 43 Knee flap attach Double needle lockstitch 2 2 2 42 44 Bar tacking Bartack machine 2 2 3 43 45 Back rise overlocks 5 threads overlock 1 1 1 44 46 Back rise topstitches Double needle lockstitch 1 1 1 45 47 Big loop matching Helper 1 2 2 46 48 Big loop runstitch Single needle lockstitch 3 3 3 47 49 Big loop turning Turning machine 2 2 2 48 50 Big loop runstitch Single needle lockstitch 2 2 2 49 51 Big loop button hole Button hole machine 1 1 1 50 52 Small loop runstitch Loop stitch machine 1 1 1 51 53 Small loop, big loop and Single needle lockstitch 3 4 4 waistband attach 52 54 Waistband topstitch Single needle lockstitch 2 2 2 53 55 Waist band closing with size Single needle lockstitch 2 2 2 and label tags 54 56 Inseam Overlock 5 threads overlock 2 2 2 55 57 Trouser turning Helper 1 1 1 56 58 Inseam topstitch Feed of arm 2 2 2 57 59 Button hole on Hip band Button hole machine 1 1 1 58 60 Button hole on the bottom leg Button hole machine 1 1 1 59 61 Bottom rope attach Helper 1 2 2 60 62 Bottom hemming Single needle lockstitch 2 2 2 61 63 Small loop tacking Single needle lockstitch 2 2 2 62 64 Final Bar tacking Bartack machine 2 2 2 63 65 Adjustable rope cutting Helper 1 1 1 64 66 Adjustable hemming Single needle lockstitch 1 1 1 65 67 1st adjustable rope attach Single needle lockstitch 1 1 2 66 68 2nd adjustable rope attach Single needle lockstitch 1 1 2 67 69 Button point marking Helper 1 1 2 68 70 Trimming Helper 7 8 8 69 71 Quality checking Quality personnel 2 2 2 70 72 Rework Single needle lockstitch 1 1 1 WSN= workstation number, OPS= operation sequence 20