Line Balancing Analysis in Ribbed Smoked Sheet Production Using Heuristics Methods (Study at PT. Wabin Jayatama

. Production line balancing is done to increase line efficiency. Product flow between workstations is stated to be better when the percentage of line efficiency is higher. One of the factors that can hinder the smooth running of a line is a bottleneck. The occurrence of a bottleneck at the Ribbed Smoked Sheet production station, namely the milling and sorting station, at PT. Wabin Jayatama caused the production line to become unbalanced. The output that will be generated from this research is the efficiency level of the actual production line and the proposed design of a production line with more optimal efficiency, which will later be used as a consideration for recommendations to improve the performance of the company's production line. This research was carried out using the Kilbridge & Wester Heuristic method and the Ranked Positional Weights Heuristic method. The results of the line balance analysis showed that the production line had a higher line efficiency (LE) of 50.05%, an increase by 25.62% to 75.67%. The initial line balance delay (BD) of 49.55% can be reduced by 25.22% to 24.33%. Then the Smoothest Index (SI) of 447.74 can decrease by 360.51 to 87.23.


Introduction
Rubber latex is one of the abundant natural resources in Indonesia.In 2014-2018, Indonesia ranked 2nd as a rubber producing and exporting country in the world.Rank 1 is occupied by Thailand.In 2018, the area of rubber plantations in Indonesia was 3,671,700 hectares, then increased in 2019, to 3,683,500 hectares [1].The largest rubber plantation area in 2019 was in the province of South Sumatra, covering an area of 861,600 hectares, while the area of rubber plantations in Banten province was 18,000 hectares.National rubber latex production in 2018 was 3,630,400 tons then decreased to 3,449,000 tons in 2019 [2].Rubber latex production in Banten province also experienced a decline in 2018-2019, successively 17,600 tons to 14,900 tons [3].In 2021, Natural rubber production in Indonesia reached 3,120,000 tons.This number increased by 8.2% compared to 2020 which amounted to 2,880,000 tons.In 2021 Indonesia will become the world's largest rubber producer and exporter [4].Potential non-food markets that will absorb rubber marketing include the automotive, tire and other industries.
Ribbed Smoked Sheet Grade I refers to rubber sheets that have been smoked, cleaned, exhibit a clear color, do not adhere to each other, are free from air bubbles, devoid of blemishes (fungus), and are free from the consequences of incomplete processing.Ribbed smoked sheet grade I production output of PT.Wabin Jayatama in January 2023 amounted to 929 kg, while the production target set by the company to meet the established demand for that month reached 1,200 kg.The high demand for ribbed smoked sheet products has made PT.Wabin Jayatama needs to optimize its production performance so that the target of Grade I Ribbed Smoked Sheet products can be achieved.Various efforts have been made by the company to optimize line performance and increase productivity, but have not paid attention to line efficiency in the process of making RSS products.Meanwhile, without realizing it, the existence of a bottleneck at the milling and sorting work stations of ribbed smoked sheets can slow down the overall processing time [5].Observations conducted for a minimum of 18 days yielded results indicating the accumulation of materials at the milling station and sorting station consecutively, during the processing of 1341 kg of raw rubber sheets and 971 kg of ribbed smoked sheets, with consecutive processing times of 106.33 minutes and 166.17 minutes, respectively.
A good production line balance is a solution to increase line efficiency.The efficiency of the production line between work stations has an effect on supporting the smooth production process.The higher the efficiency percentage of the line, the better the product flow between work stations.The method to be used in this study is the Kilbridge & Wester method and Ranked Positional Weights with 3 parameters, namely Line Efficiency (LE), Balance Delay (BD), and Smoothest Index (SI).Line efficiency is a measure of how well a production line utilizes its resources to meet production targets, often expressed as a percentage of actual output relative to the maximum possible output.Balance delay refers to the time lost in a production line due to imbalances in workloads among different stations or processes, leading to inefficiencies and slower overall production.The Smoothness Index quantifies the level of consistency and regularity in the production process, indicating the degree to which variations or disruptions are minimized for smoother and more reliable manufacturing operations.The purpose of balancing production lines in the rubber industry is to maximize line efficiency, minimize the number of work stations, and reduce the possibility of bottlenecks on the production line.A balanced production trajectory has the advantage of a smooth production process.The smoothness of the production line must be maintained in order to create a balance in the production line of Ribbed Smoked Sheet grade I PT.Wabin Jayatama.

Research methods
This research was conducted to determine the line efficiency, the optimal number of work stations, and the optimal processing time in grade I ribbed smoked sheet division at PT Wabin Jayatama.In general, the research procedure can be seen in Figure 1.There are 2 types of data used in this study, namely primary and secondary data.The primary data needed in this study is the operating time during working hours using the stopwatch time study method.While the secondary data used in this study is the history, profile, and organizational structure of the company.

Data testing
A data uniformity test was carried out to ensure that the data collected in the form of an The data adequacy test was conducted to determine whether the amount of observed data was sufficient to conduct research.The data adequacy test formula is as follows [7]:

Calculation of Cycle Time (Ct)
The calculation of cycle time for each work station is obtained by the following formula [8]:

Creating precedence diagrams
Precedence diagrams are made to show the order of the packaging process and the time in each production process of grade I ribbed smoked sheet at PT. Wabin Jayatama.

Line Effecieny
Line efficiency is the efficiency level of the production line expressed as a percentage.
The formula for calculating line efficiency is as follows [9]:

Balance Delay
Balance Delay is a measure of the inefficiency of a path caused by imperfect allocation of work stations.Balance delay is obtained by the following formula [9]:

Smoothest Index
Line efficiency is the efficiency level of the production line expressed as a percentage.

2.2.4
Analysis with the Kilbridge-Wester Heuristic Method a. Creates a priority chart so that nodes representing work elements with identical priorities are arranged vertically in columns.b.Enters elements in column order, column I at the top of the list.If an element can be placed in more than one column, list all columns by element to indicate the transferability of that element.c.To assign an element to a workstation, start with the column element

Results and discussion
Fig.This study takes the object of the production process of smoked rubber sheets at PT. Wabin Jayatama.The data collection and research was conducted over a three-month period, commencing from January 2023 to March 2023.The thing that needs to be done in carrying out research on the balance of production lines is to map the operation process planning chart (Operation Planning Chart / OPC).In the production of smoked rubber sheets, there are 11 operational activities, and two joint activities (operations and inspections).

Data uniformity test
The data uniformity test is carried out to ensure that the data collected in the form of an average productive percentage is within the control limits.Testing the data in this study uses a confidence level of 95% with a value of k (an index whose size depends on the level of confidence chosen) of 2 and an accuracy level of 5%.The recapitulation of the results of the data uniformity test can be seen in Table 1.  1 Shows the upper control limits (UCL) and lower control limits (LCL) for each packaging operator.As a result, data on the percentage of productive activity for all operators from day 1 to day 15 are within the range of the upper and lower control limits.This shows that the average productive percentage of all operators is uniform.Hence it can be said that the packaging process is in control.According to [10], data uniformity is characterized by no data that is out of control, that is, data that is outside the range of control limits.The quantity of data collected during the observation of Grade I ribbed smoked sheet production was 18 sets of 8 data points each, with a minimum observation period of 18 days, resulting in a total of 144 data points for the observational dataset (N).The recapitulation of the results of the packaging operator data adequacy test can be seen in Table 2. of data adequacy (N').[11] stated that if the total number of observed data (N) is greater than the amount of data adequacy (N'), then the data is sufficient and can be used for research.If the data is insufficient, observations must be continued until the amount of data is sufficient.

Actual line balance analysis
Analysis of the actual line balance is carried out to explain the state of the line balance in actual conditions.The analysis begins by making the precedence diagram shown in Fig. 2.This is in accordance with the statement of [12], which states that the precedence diagram is used before entering the completion stage of the line balance to facilitate planning.After describing the packaging path systematically with precedence diagrams, the next step is to analyze the path balance.The data used are cycle time and standard time data from each operator, as well as the number of work stations.The results of the analysis of the balance of the actual packaging path can be seen in Table 4.

Cycle time calculation
Cycle time, namely the maximum working time limit of each workstation.The maximum time in Table 3 is 166.17 minutes.Since C ≥ t max , C = cycle time; t max = the greatest time of all work elements; t max = 166.17minutes ≈ 167 minutes, so the cycle time is 167 minutes.

Preparation of proposed precedence diagram design
Based on the initial conditions of the company, it is known that the company has not implemented line balancing in the Grade I Ribbed Smoked Sheet Factory division.In principle, the Kilbridge & Wester method and the Ranked Positional Weights (RPW) method try to assign tasks first to operations that have precedence.The first step that needs to be done before doing the calculation is to set the precedence constraints.Precedence constraints are predecessor constraints in completing a work element, there are sequences of tasks that must be fulfilled beforehand so that the element can be executed [13].Activities start from element O-6 because activities O-1 to O-5 have started the previous day (work in process).The making of this precedence diagram is in accordance with the statement of [14] stated that the precedence diagram is used before entering the completion stage of the line balance to facilitate planning.The circle symbol indicates work stations, the arrow symbol indicates the dependency of the process, and the number above the circle symbol represents the time.The Precedence Diagram can be seen in Fig. 3 below.Fig. 3 shows the proposed precedence diagram design which has been divided into two paths and 8 regions.The red circle symbol indicates that the activity is carried out by group A. The blue circle symbol indicates the activity is carried out by group B. The yellow circle symbol indicates that the activity is carried out by group C. The green circle symbol indicates the activity is carried out by Group B and C workers (proposal) so that workers B and C do not have idle time.Each group represents distinct workforce with workloads tailored to individual workers and activities adjusted according to worker's skills.For instance, in the red-colored work group, there is a composition of both male and female workers, while Work Groups B and C have significant workloads, leading to the inclusion of only male workers.

Applying calculations using Kilbridge and Wester method
After obtaining the theoretical number of regions (8 regions), it is possible to group work stations based on their tasks.The total in one station must be less than the cycle time (167 minutes), while the efficiency is taken from (the amount of time at each work station/maximum time) and is expressed as a percentage.This statement is supported by [15] cycle time, namely the maximum working time limit of each workstation.The division of work stations using the Kilbridge & Wester method can be seen in Table 5.

Calculations with the Ranked Positional Weight (RPW) method
Line balancing using the Ranked Positional Weight (RPW) method is then followed by calculating the position weight of each work station based on the total operating time and the following operations.If the position weights are available, the elements of operating activities can be arranged based on their ranking.After weighing the positions for each operation, the next step is to determine the composition of the work stations after they have been balanced using the Ranked Positional Weights method.The RPW value obtained in Table 6 is the value obtained from the weight in minutes that has been calculated previously.The RPW values obtained are then sorted and a ranking assessment is carried out.The proposed trajectory design submitted to PT. Wabin Jayatama has two lines because the RSS sorting does not depend on washing the rubber sheets on the same day.

Improvement recommendations
The Kilbridge & Wester method and Ranked Positional Weights that have been carried out will then be compared with the initial trajectory using three parameters, namely Line Efficiency (LE), Balance Delay (BD) and Smoothest Index (SI).In Table 7 below is a comparison of the performance of the production line balance on the initial line with the results of calculations using the Kilbridge & Wester method and Ranked Positional Weights.Ranked Positioned Weight method.Under current conditions, the line efficiency obtained was only 50.05%, but after improvements were made, the line efficiency increased by 25.62% to 75.67%.The increase in line efficiency is due to the effective processing of each operation at each work station.With the increase in line efficiency, it causes a decrease in Balance Delay (BD) of 25.22% from 49.55% to 24.33%.The workload balance that occurred resulted in a better Smoothness Index (SI), namely from 477.74, it decreased by 360.51 to 87.23.By looking at the above performance, with the line balancing method, the production process can be more efficient, so as to increase the productivity of the company.The division of work stations can be seen in Fig. 4.

Managerial implications
The implication of this research for the company is to rearrange the groups of workers so that workers do not have excess idle time and work does not pile up at one station.The suggestion from the results of this study is to use Group B and C personnel to do one of Group A's jobs, namely sorting when Groups B and C don't have a job.Moving the sorting work elements to the beginning that was done by group B and C workers also had a big impact on the time of activities per day.The daily production time of 521.37 minutes can be reduced to 443.17 minutes, optimizing production time.The company can enhance production capacity to yield greater profits once it reaches the company's target.The application of the line balancing concept using the Kilbridge & Wester method and Ranked Positional Weight only takes the aspect of operating process time and the sequence of interrelationships between operating processes.It is necessary to pay attention to the assignment of workers to new stations related to the level of suitability of the skills of the workforce concerned with the new job.This is because these two methods do not consider other aspects such as the type of work, operator capability, working environment conditions, level of product accuracy, type of production machine/equipment, and product quality.Further research can be carried out so that it can provide more benefits for the company.

Conclusions
Based on the analysis of the balance of the PT.Wabin Jayatama, the production line for ribbed smoked sheets has a low Line Efficiency (LE) value of 50.05% with a Balance Delay (BD) value of 49.55%.Meanwhile, the Smoothest Index (SI) value on the actual ribbed smoked sheet production line is very high, namely 447.74.The low efficiency value of grade I ribbed smoked sheet production line is caused by a bottleneck at the milling and sorting work stations, so that the flow of the production process is not smooth.The flow of the production process will be smoother when line balancing is carried out.
The results of the calculation of the balance of the production line of smoked rubber sheets were carried out using the Kilbridge & Wester method and the Ranked Positional Weights method.The results of the track balance analysis showed that the production line of grade I smoked rubber sheets had a higher line efficiency (LE) of 50.05%, an increase of 25.62% to 75.67%.Initial track balance delay (BD) of 49.55% can be reduced by 25.22% to 24.33%.Then the Smoothest Index (SI) of 447.74 could decrease by 360.51 to 87.23.

2. 2 . 5
Analysis with Ranked Positional Weights (RPW) Method a. Calculates the RPW for each element by adding the element times together with the times for all following elements in the arrow chain of the precedence diagram.b.Lists elements in order of their RPW, the largest RPW at the top of the list.c.Assign elements to stations according to RPW, avoid precedence diagram limitations and cycle time violations.

Fig. 2
Fig. 2 Precedence diagram of RSS grade I production.

Fig. 4 .
Fig. 4. The division of the Kilbridge and RPW Method Work Stations

Table 2 .
Data adequacy test results.

Table 2
, all operators have a larger amount of observed data (N) than the amount

Table 3 .
Distribution of initial track work stations (unbalanced).

Table 4 .
Results of actual track balance analysis.

Table 5 .
Distribution of work stations using the Kilbridge & Wester method.

Table 6 .
Distribution of RPW method stations with track adjustments.

Table 7 .
Results performance recommendations for improvement.
Line balance methods that can be considered are the Kilbridge & Wester method and the