Identification and Minimization of Waste with the Application of Lean Manufacturing in the Cireng Production Process at PT Brecxelle Berkah Uniti

. Based on the results of observations and the 7 waste questionnaires at PT Brecxelle Berkah Uniti production, it was found that the waste of defects occurs such as cireng are shapeless, destroyed, and do not pass through the metal detector machine. Waste of waiting occurs during the cooling process, when starting the engine, and waiting for the pressing process. Waste of transportation occurs due to circular and non-consecutive process flows. This research aims to minimize production time and increase production result by simulating the cireng production process of PT Brecxelle Berkah Uniti using the Lean Manufacturing and System Simulation method. Based on the simulation results with the proposed improvements, the results obtained from the simulation were that the total production lead time was reduced by 10,97 minutes from 40.45 minutes to 29,48 minutes. The NVA and NNVA time on FVSM also decreased by 10.81 minutes from the activity time on CVSM. Based on the simulation results, production results increased by 80% after the proposed improvements, from 4,671,945 pcs per month to 8.429.679 pcs per month.


Introduction
PT Brecxelle Berkah Uniti, this company needs to pay more attention to each production line.Based on the percentage of defects in products produced by PT Brecxelle Berkah Uniti, there were the highest defects in April and January at 0.156% because these months coincided with the new year holidays and the month of Ramadan so the demand for cireng products also increased.Based on the results of interviews with Quality Control staff, the percentage of these defects has exceeded the company's target of 0.050%.
In addition to waste defects, there is waste transportation as a result of rotating and nonconsecutive process flows, resulting in waste of transportation between cooling work stations and expiration date manufacturing work stations.In addition, workers are required to take the dough and return the tray to the room they came from.This has an impact on increasing worker transportation time so that production time becomes longer.Based on processing time data, a total transportation time of 134.427 seconds is obtained with a distance of 42 m so the distance is expected to be reduced and transportation time can be reduced.
Therefore, to reduce waste, a proposed improvement is needed to maintain quality and increase productivity in the cireng production process.With this research, it can be seen how the proposed improvements can minimize waste with the Lean Manufacturing method and know the output of the proposed improvements with the System Simulation approach.

Research Method
The research was conducted by examining waste at PT Brecxelle Berkah Uniti, starting with the preparation phase, the data collection phase, as well as the data processing and analysis phase.appropriate place in the text.

Preparation
The preparation phase includes field studies and literature studies related to Lean Manufacturing, production processes, 7 Waste, Value Stream Mapping (VSM), Value Stream Analysis Tools (VALSAT), 5 Whys, and System Simulation to formulate problems and determine objectives and benefits of research.

Data Collection
You Data collection methods that used in this research include interviews, questionnaires, observation, and literature study.The types of data and procedures used for data collection in this study are described as follows:

Primary Data
Primary data contained from observation, interview and questionnaire.The data obtained from the observations is data regarding conditions that related to the research problem, such as the flow production process, layout, and transportation flow during the production process.The data obtained from the interview results are data on the sequence of production activities, the number of operators for each production activity, the distance of each workstation, the working hours of all workers, and the target defect of the company.The data obtained from the results of the questionnaire is the score or weight data from the 7 waste questionnaire.

Secondary Data
In this study, the secondary data used is production data and defect products data by PT Brecxelle in July 2021 -August 2022.The researchers obtained this data from the company's internal data for 2021 and 2022.

Score Results and Waste Rating from the Questionnaire
The following is the result of data recapitulation and the results of the 7 Waste Questionnaire: Based on the questionnaire above, by taking the 3 wastes with the highest percentage of waste, improvements can be made for 68% of the waste in the company or more than half of the waste in the company.It means more than half of the waste will be identified and proposed improvements.

Value Stream Analysis Tools (VALSAT)
The following is the result of the VALSAT calculation by multiplying the average waste with the multiplier waste correlation value according to the VALSAT table.

Table 3. Calculation of Value Stream Analysis Tools
Based on the calculation above, it is known that Process Activity Mapping is the selected tool that has the highest score, which is 37,8.

Actual Process Activity Mapping (PAM)
Based on the actual Process Activity Mapping table, it can be seen the total number of all types of activities (Operation, Transportation, Inspection, Storage, Delay), activity time, and activity categories (Value Added, Necessary Non-Value Added, Non-Value Added).Here is the recapitulation:

Root Cause Analysis with 5 Whys
Based on the three waste that have the highest score of 7 waste questionnaire, the following is a root cause analysis using the 5 Whys method to identify the root causes of the PT Brexcelle Berkah Uniti production unit: 1. Defects Operators are chasing time so that there is no delay in the molding process Why 4 The molding process is still manual with human hands so the process takes a long time Why 5 There is no machine that is suitable for molding cireng in the shape desired by the company and the company does not have a standard diameter or size for cireng with defects Why 1 The cireng dough is too thin Why 2 Operators in rush when molding the dough Why 3 Operators are chasing time so that there is no delay in the molding process Why 4 The molding process is still manual with human hands so the process takes a long time Why 5 There is no machine that is suitable for molding cireng in the shape desired by the company and the company does not have a standard diameter or size for cireng with defects Why 1 There is contamination of the production from operator's pins/needles as well as pieces of iron or wire from the flour sieves during the molding process Why 2 Lack of carefulness from the operator that they did not realize that there was pins/needles Why 3 Operators are not aware of the dangers of using pins/needles in the food industry and pieces of wire that are not visible to the eye   There are not many container, so the container must be returned to the room where they came from, thus increasing the transportation time

Why 3
The distance between stations is quite far and there are no resources to carry the goods

Why 4 None operator collects containers from the molding table
Why 5 There is no tool provided by the company to be able to deliver the container back to the room of origin The dough can be placed on a cooling rack so that the cooling process can be more even.
The exhaust fan is used to suck in and throw away the hot air that is on the rack so that the hot air will be sucked out.b.Waiting for Pressing Process Add 2 machines so that the number of machines becomes 6 machines.This is done to reduce the waiting time for the press machine.

Process Activity Mapping (PAM) Improvements
The following are some changes in production time based on proposed improvements and Process Activity Mapping Improvements to the production unit of PT Brecxelle Berkah Uniti.

Table 9. Improvement in Process Time
Based on the improvement Process Activity Mapping table, it is known that there are several changes in activity time and reductions in activities without added value (Non-Value Added).

Table 10. Comparison of Actual and Proposed Production Time
With the proposed improvements, the production time for each production process is reduced.Total production lead time was reduced by 657.97 seconds or 10.97 minutes from 2426.77 seconds or 40.45 minutes to 1768.80 seconds or 29.48 minutes.

Table 11. Comparison of Actual and Proposed Activity Time
Based on the table above, it can be seen from the results of the proposed improvements, it is obtained a reduction of 5 activities with details of 1 operation activity and 4 transportation activities.So with the proposed improvements, there are 34 activities with details of 16 Operations, 13 Transportation, 2 Inspections, and 3 Delays.From these data, it is known that the dominant activities are Operations and Transportation.

Actual Simulation Model Design
Before do the simulation, it is necessary to do the distribution test to determine the best distribution that will be used in the simulation modeling with the Stat::Fit tool contained in the ProModel Software.The following is the difference in the simulation output in 1 month which was replicated 12 times.

Table 13. Comparison of Real Total Production with Simulation Results
Wilcoxon test was performed with Minitab software.Below is the basis decision-making for the Wilcoxon test: H0 : There is no difference between the actual simulation results and the real system results H1 : There is a difference between the actual simulation results and the real system results If the P-Value > 0.05, then H0 is accepted

Fig. 7. Wilcoxon Test of Actual Simulation
Based on the Wilcoxon Test table, obtained P-Value which is 0.610 where the P-Value has exceeded the α value of 0.05, so H0 is acceptable.H0 is accepted, meaning that there is no difference between the actual simulation results and the real system results.So, based on the results of the Wilcoxon test, the model simulation is validated because there is no difference between the actual simulation results and the real system results.

Simulation Model Design Improvements
Improvement for simulation model design is made in order to see the overall impact of changes to the actual process and proposed improvements.Based on the Wilcoxon Test table, it was found that the P-Value was 0.003 where the P-Value was less than the α value, which was 0.05, it means H0 is rejected.
H0 is rejected, meaning there is a difference between the simulation results based on proposed improvements and the results of the real system.So, based on the results of the Wilcoxon Test, the proposed improvements have an effect on the proposed model simulation so that production increases compared to the production results of the real system.

CVSM and FVSM Difference Analysis
The following is the difference in the total time of Value Added, Non-Value Added, and Necessary Non-Value Added activities on CVSM and FVSM: Based on the differences in FVSM and CVSM, it can be seen that in the cireng production process of PT Brecxelle Berkah Uniti, companies can reduce lead time or speed up activity time if they do the proposed improvements.

Analysis of Differences in Simulation Results
The total results of the improvement simulation production are compared with the actual production results of the real system at the company to see any differences in production results after the repairs are carried out.Based on the table above, it can be seen that with the same time of 1 month and 192 working hours, the proposed simulation production results are higher than the actual system production results in real conditions.The average difference in the proposed simulation production results with the real system is 3,757,734 pcs.
The increase in production from the previous 4,671,945 pcs to 8,429,679 pcs per month was due to proposed improvements in the form of reducing the time in the production process.These improvements

Fig. 2 .
Fig. 2. Current Value Stream Mapping a. Waiting for Cooling Down Process b.Waiting for Press Process 3. Excessive Transportation a. Inefficient Material Handling for Packing Process from Cooling Workstation to Packing Workstation.b.Excessive Transportation when Taking the Dough and Returning the Tray to the Origin Room.

Fig. 3 .
Fig. 3. Actual Production Layout of PT Brecxelle Berkah Uniti Why 4There is no socialization and prohibition on the use of safety pins/needles in the production process Why 5There is no SOP regarding the use of pins/needles during machine to divide the dough into 15 grams per piece of cireng and a dough sheeter machine to flatten the dough into 1 cm thick.c.Product Does Not Pass the Metal Detector Machine Making written SOPs (Standard Operating Procedures) regarding the use of safety pins, and other sharp objects in the production unit so that no sharp items accidentally enter the food processing.2. Waiting a. Waiting for Cooling Down Process Procure 2 stacks of cooling racks and exhaust fans to cool down the product in a short time.
3. Excessive Transportation a. Inefficient Material Handling for Packing Process from Cooling Workstation to Packing Workstation.Set the layout of the cooling, packing, and expiration date work stations to be directional and sequential; placing chili sauce on each operator table; as well as moving the work station for making expiration dates to be between the cooling work station and the packing work station.b.Excessive Transportation when Taking the Dough and Returning the Tray to the Origin Room.Set the layout of the divider machine closer to the dough room, and set the layout of the dough sheeter machine closer to the divider machine.

Table 1 .
Output Production Data

Table 2 .
Summary of Results Score 7 Waste Questionnaire

Table 4 .
Recapitulation of Actual Activity TimeBased on the table above, it can be seen that there are 39 activities with a total lead time of 2426,769 seconds.From the table, it is obtained that the domination of activity time is delayed activity.

Table 5 .
Recapitulation of Actual Activity Time

Table 8 .
5 Whys Analysis Waste Excessive Transportation Table capacity to accommodate cireng is inadequate The table that made of stainless steel so that the heat from the bread is sent back by the table frame Why 4 Cooling table that does not have good air circulation Why 5 The cooling table has no cavity for air circulation

Table 12 .
Tool Stat::Fit Actual Simulation Distribution Results

Table 14 .
Improvements of Tool Stat::Fit Distribution ResultA performance test is conducted to find out whether there is a difference between the actual production results and the simulation production results after improvements are done.

Table 15 .
Comparison of Real Total Production with Simulation Results Improvements

Table 16 .
Recapitulation of VA Activity Time Differences in CVSM and FVSM Based on the table above, it was found that the activity time for Value Added on FVSM was reduced by 9.43 seconds or 0.16 minutes compared to the activity time for Value Added on CVSM.

Table 17 .
Recapitulation of NVA and NNVA Activity Time Differences in CVSM and FVSM Based on the table above, it was found that the activity time for Non-Value Added and Necessary Non-Value Added on FVSM decreased by 648.53 seconds or 10.81 minutes from the time on CVSM.

Table 18 .
Recapitulation of Total Actual Production and Improvement Simulation (Pcs)