Enhancing Overall Equipment Effectiveness in Indonesian Automotive SMEs: A TPM Approach

3.1 Result of data collection

3.1.1 Working hour of blowing machine

Working hour is the overall time that shows the number of working hours used for the production process. RMA Ltd operates for 5 working days a week, this research was conducted only on shift 1. In September and November 2022 there was overtime, each month 2 times overtime. The overtime time is 1 hour 30 minutes. The working hour data of the blowing machine in September 2022-November 2022 can be seen in Table 1.

Table 1. Working hour data of blowing machine

3.1.2 Planned downtime of blowing machine

Planned downtime is data showing the machine downtime the company has planned for scheduled maintenance or other management activities. Routine activities before carrying out the production process at RMA Ltd are gymnastics and briefing, then continued with the preparation time for operator preparation to run the machine after the production process has been completed, followed by cleaning activities to clean the work area and scrap from the production process. The scheduled time may change or not be the same and is flexible according to needs. The planned downtime data of the RMA Ltd blowing machine can be seen in Table 2.

3.1.3 Breakdown time of blowing machine

Breakdown time is data that shows the time of disruption to the machine so that the machine must be repaired immediately, even though the machine is operating. Then, the machine stops operating for a while. RMA LTD blowing machine breakdown time data can be seen in Table 3.

3.1.4 Setup and adjustment time of blowing machine

Setup and adjustment time shows the time it takes for the machine to start operating to produce new components, such as setting up a machine or changing moulds or dies. RMA Ltd's blowing machine setup and adjustment data can be seen in Table 4.

Table 2. Planned downtime of blowing machine

Table 3. Breakdown time of blowing machine

Table 4. Setup and adjustment data of blowing machine

In the production process, products are good, and some products have defects that must be rejected or repaired. The production data of the RMA LTD blowing machine can be seen in Table 5. In addition, the performance of the blowing machine (Table 6) also needs to be considered to assess the ideal output produced.

Table 5. Production data of blowing machine

Table 6. Performance data of blowing machine

3.2 Data processing

The data collected during the research at RMA LTD will then be processed, namely calculating the availability rate, performance efficiency, quality rate, OEE, and six significant losses carried out on the blowing machine. Availability rate is a ratio that shows the utilisation of time available for machine or equipment operation activities expressed as a percentage, for example, the calculation of availability rate in September 2022 (Eq. (1)). The availability rate percentage for September 2022-November 2022 can be seen in Table 7.

$\begin{gathered}\text { Availability Rate }=\frac{\text { Operating Time }}{\text { Loading Time }} \times 100 \% =\frac{10,900-1247}{10,900} \times 100 \%=88.56 \%\end{gathered}$         (1)

Table 7. Calculation of availability rate

Performance efficiency is a ratio that shows the ability of equipment or machinery to produce products expressed in percentages, for example, the calculation of performance efficiency in September 2022 (Eq. (2)). The percentage of performance efficiency for September 2022-November 2022 can be seen in Table 8.

$\begin{gathered}\text { Performance }= \frac{\text { Processed Amount } \times \text { Ideal Cycle Time }}{\text { Operating time }} \times 100 \% \\ =\frac{58,811 \times 0.12}{9,653} \times 100 \%=73.11 \%\end{gathered}$        (2)

Table 8. Calculation of performance efficiency

The quality rate calculation is a ratio that shows the ability of a machine to produce products that meet standards and is expressed as a percentage, for example, the calculation of the quality rate in September 2022 (Eq. (3)). The quality rate percentage for September 2022-November 2022 can be seen in Table 9.

$\begin{gathered}\text { Quality Rate }& =\frac{\text { Processed Amount }- \text { Defect Amount }}{\text { Processed Amount }} \times 100 \% \\ &=\frac{58,811-(589+257)}{58,811} \times 100 \%=98.56 \%\end{gathered}$            (3)

Table 9. Calculation of quality rate

The OEE calculation is based on the availability, performance, and quality values obtained; for example, the OEE calculation in September 2022 (Eq. (4)). The OEE percentage for September 2022-November 2022 can be seen in Table 10.

$\begin{gathered}\mathrm{OEE}&=\text { Availability } \text { Rate } \times \text { Performance Efficiency } \times \text { Quality } \text { Rate } \\ &=88,56 \% \times 73,11 \% \times 98,56 =63,81 \%\end{gathered}$        (4)

Table 10. Calculation of OEE

The calculation of six big losses aims to see what losses occur in the production process, resulting in less effectiveness. The calculation of six big losses consists of breakdown losses, set-up and adjustment losses, reduced speed losses, idling and minor stoppages, rework losses, and reduced yield.

Breakdown losses are a visible cause of loss because the damage will impact the machine, which will not produce output, for example, the calculation of breakdown losses in September 2022 (Eq. (5)). The percentage of breakdown losses for September 2022-November 2022 can be seen in Table 11.

$\begin{gathered}\text { Breakdown Losses }=\frac{\text { Total Breakdown Time }}{\text { Loading Time }} \times 100 \% \\ =\frac{729}{10.900} \times 100 \%=6.69 \%\end{gathered}$        (5)

Table 11. Calculation of breakdown losses

Setup and adjustment losses are caused by the preparation of a long production process due to waiting for the arrival of materials and machine settings, for example, the calculation of setup and adjustment losses in September 2022 (Eq. (6)). The percentage of setup and adjustment losses for September 2022-November 2022 can be seen in Table 12.

$\begin{aligned} \text { Setup & Adjust Losses} &=\frac{\text { Total Setup & Adjustment Time }}{\text { Loading Time }} \times 100 \% \\ & =\frac{518}{10,900} \times 100 \%=4.75 \% \\ & \end{aligned}$           (6)

Table 12. Calculation of setup and adjustment losses

Reduced speed losses are caused by a decrease in engine speed in carrying out its operations or the engine does not work optimally, for example, the calculation of reduced speed losses in September 2022 (Eq. (7)). The percentage of reduced speed losses for September 2022-November 2022 can be seen in Table 13.

$\begin{aligned}\text { Operating Time (Ideal Cycle Time } \times \text { Result Processed) } \text { Loading Time } \times 100 \%\\= \frac{9,959-(0.12 \times 58,811)}{10,900} \times 100 \%=26.62 \% \\ & \end{aligned}$            (7)

Table 13. Calculation of reduce speed losses

Idling and minor stoppages cause the production process to stop for no more than five minutes but with a fairly frequent stoppage frequency, which can also be called machine idle time or non-productive time; for example, the calculation of idling and minor stoppages in September 2022 (Eq. (8)). The percentage of idling and minor stoppages for September 2022-November 2022 can be seen in Table 14.

$\begin{aligned} (\text { Total Target } \times \text { Total Production }) \times \text { Ideal Cycle Time } \text { Loading Time } \times 100 \%\\=\frac{(71,675 \times 58,811) \times 0.12}{10,900}\times 100 \%=14.16 \% \\ & \end{aligned}$       (8)

Table 14. Calculation of idling and minor stoppages

Rework losses are caused by the production process producing defective products, and the defective products produced result in material losses, reduced quantities that have been produced, increased production waste, and require costs for rework, for example, the calculation of rework losses in September 2022 (Eq. (9)). The percentage of rework losses for September 2022-November 2022 can be seen in Table 15.

$\begin{aligned} \text { Rework Losses } & =\frac{\text { Ideal Cycle Time } \times \text { Product Rework }}{\text { Loading Time }} \times 100 \% \\&=\frac{0.12 \times 257}{10,900} \times 100 \%=0.28 \%\end{aligned}$              (9)

Table 15. Calculation of rework losses

Reduced yield is a loss that arises during the production process, resulting in products that do not meet the standards due to the production process not yet reaching a stable condition, for example, the calculation of reduced yield in September 2022 (Eq. (10)). The percentage of reduced yield for September 2022-November 2022 can be seen in Table 16.

$\begin{gathered}\text { Reduced Yield }=\frac{\text { Ideal Cycle Time } \times \text { Product Scrap }}{\text { Loading Time }}\times 100 \%\\=\frac{0.12 \times 589}{10,900} \times 100 \%=0.65 \%\end{gathered}$            (10)

Table 16. Calculation of reduced yield

3.3 OEE and six big losses analysis

The Pareto chart of the six big losses can be seen in Figure 2. The calculation of the six big losses that have been accumulated shows that reduced speed losses are the largest type of losses that occur.

2.png

Figure 2. Pareto chart of six big losses

Analysis of OEE calculations is carried out in order to determine the level of effectiveness of the use of blowing machines in September 2022-November 2022 (Table 17). This OEE measurement uses average data from September 2022 to November 2022, namely by multiplying the availability value, performance efficiency value, and quality rate value.

Table 17. Percentage of OEE value

The OEE value of the calculation for three months made the average result of 66.42%, and the lowest category is performance efficiency, which is 75.54%. The explanation is as follows:

1. Availability rate data obtained from the calculation of its average of 89.42%. This value can already be categorised as quite good; what affects this value is the availability time used during operation, namely operation time and loading time, besides that the availability rate value will be a good category if the number of breakdown times can be minimised because the more damage will result in the higher the availability rate value.
2. Performance efficiency data obtained from the average calculation of 76.66%. This value cannot be categorised as good and must be increased again; what affects this value is the availability time for the process to operate the resulting output and the ideal cycle time. This calculation of the amount of output does not match the time spent during the production process, so the value obtained can be categorised as not good.
3. The quality rate data obtained from the average calculation is 98.34%. This value is categorised as good; what affects this value is the amount of production and the number of defective products. The defective parts that occur can be categorised as not affecting production results because the quality rate value is good.
4. The OEE value is obtained from the product of the three factors, a value of 66.42%. This value still needs improvement because it is far from the standard set by Japan Institute of Plant Maintenance (JIPM), which is 85%. This value can be low because performance efficiency provides a small value that impacts the OEE value. Small performance efficiency is caused by the amount of production that fails to meet the target.

Analysis of six big losses aims to determine what factors affect the low effectiveness of blowing machines. The relationship between OEE and six big losses has an inversely proportional relationship. If the OEE value is low, it will result in a high value of six big losses, and vice versa. Based on the results of the calculations that have been carried out, the factor that causes the low OEE value occurs in performance efficiency. The results of the calculation of six big losses and the percentage order for September 2022-November 2022 can be seen in Table 18.

The six big losses for three months are calculated using the average. The results obtained reduce speed losses with the highest value, affecting the value of the actual production time, actual production quantity, ideal cycle time, and loading time. The value that most affects the low reduced speed losses is the time spent not following the output obtained, and the ideal cycle time that has been determined with the actual cycle time is different and slower.

Table 18. Percentage of six big losses

The next stage is to analyse the results of the most dominant type of losses in the calculation of six big losses, namely reduced speed losses, using Fishbone diagrams in order to find out the root causes of the problems that cause the dominant losses, look for any factors that cause reduced speed losses. The factor is sought for the cause of the problem and to find out the root of the problem. Making this Fishbone diagram results from discussions with operators and maintenance staff. The fishbone diagram for reduced speed losses can be seen in Figure 3.

3.png

Figure 3. Fishbone diagram of reduce speed losses

Four factors cause reduced speed losses (Figure 3) to be the most dominant losses: humans, materials, methods, and machines. The explanation of each factor is as follows:

(1) Man

The problem is the operator makes machine setup errors. This problem happens because the operator is not careful.

(2) Material

The material factor, which is the problem, is that the recycled material needs to follow the machine's specifications. This problem happens because recycled material processing still needs to be corrected, so the plastic beans produced from recycled processing are quite large.

(3) Method

The method factor that is a problem is that machine checking and maintenance do not exist because there is no definite procedure.

(4) Machine

The machine factor that is a problem is that the replacement dynamo spare parts need to match the specifications. The previous dynamo caught fire, so the company had to look for a replacement dynamo. The production process needs to run smoothly, and the machine is reduced in speed.

After the factors that affect reducing speed losses are successfully mapped, a 5W + 1H analysis is made. 5W + 1H analysis is carried out to bring up the resolution of the problems that have been analysed in the Fishbone diagram. Each factor in the Fishbone diagram will be sought for problem-solving so that the problems can be resolved. The 5W+1H analysis of reduced speed losses can be seen in Table 19.

Table 19. 5W+1H analysis

3.4 Improvement implementation

3.4.1 Socialising and making work instructions

This improvement was made because the operator needed to set up the machine correctly. With the supervisor re-socialising the operator to do the machine setup for each type of part, the operator is expected to avoid making mistakes again. The socialisation done by the supervisor can be seen in Figure 4.

4.png

Figure 4. Socialisation by supervisor

Making work instructions is the next stage so machine setup errors do not occur again. The goal is for operators to see the procedures and rules for each type of cover door for machine setup. The impact is that operators find it easier to do machine setup so that the goods produced follow the standards and the course of the production process follows the target. Based on the data, the number of operators making machine setup errors in September 2022-November 2022 occurred 3 times, and after the improvements seen in December 2022-February 2023, there were no such errors. In the work instructions, there are standard blowing machine spare parts so that there are no errors in purchasing and installing machine spare parts. Work instructions for the blowing machine can be seen in Figure 5.

3.4.2 Material standardisation

This improvement is done so that operators pay attention to the material used in the machine. Materials that do not follow the standard will quickly damage the filter part of the machine, adding machine breakdown time to repair the filter so that the production process time will be disrupted. The use of recycled materials for the subcover consists of 20%-25% original plastic seeds, the rest recycled plastic seeds, and 100% recycled plastic seeds for plastic packing.

Improvements were made because the size of the recycled material was enormous, so the machine filter was damaged twice a month. After regular checks and paying attention to the standard size of the material, damage to the machine filter occurred once a month. This recycled material aims to reduce costs, so its characteristics must be improved for the size to be appropriate, as in Figure 6.

5.png

Figure 5. Example of work instruction for blowing machine (in Bahasa)

6.png

Figure 6. Material standard of blowing machine

3.4.3 Improvement of check sheet creation

This improvement is carried out in the hope that the blowing machine will not easily experience breakdowns and that production targets are achieved if the problem can be resolved, it will have an impact on the availability value and performance efficiency value because it is related to the breakdown time and the amount of production produced, before implementing the check sheet, the calculation of Mean Time To Repair (MTTR), Mean Time Before Failure (MTBF), Mean Time To Failure (MTTF) is carried out.

$\begin{gathered}\text { MTTR }=\frac{\text { Total Downtime }}{\text { Number of Failures }} \times 100 \% \\ =\frac{1,779}{14} \times 100 \%=127.07 \text { minutes }\end{gathered}$              (11)

Based on the MTTR calculation (Eq. (11)) that has been carried out, the average time required for repair when damage or problems occur is 127.07 minutes.

$\begin{gathered}\text { MTTF }=\frac{\text { Total Downtime }}{\text { Number of Failures }} \times 100 \% \\ =\frac{28,745}{13} \times 100 \%=2,211.15 \text { minutes }\end{gathered}$          (12)

Based on the MTTF calculations that have been carried out, the average time the machine can operate before the subsequent damage occurs is 2,211.15 minutes, in the case of the dynamo checked for 3 months. The replacement of the dynamo is carried out for 5 years due to the infrequent frequency of damage, for the engine filter is replaced for 2 weeks once due to the frequency of damage 1 month 2 times.

$\begin{gathered}\text { MTBF }=\frac{\text { Total Downtime }}{\text { Number of Failures }} \times 100 \% \\ =\frac{28.745}{1}=28,745 \text { minutes }\end{gathered}$           (13)

Based on the MTBF calculations that have been carried out, the average time the machine can operate before the next damage occurs is 28,745 minutes.

3.5 Impact of improvement

3.5.1 OEE after improvement

The OEE value is calculated again after the improvement implementation. The OEE value is expected to change for the better because the increase in the OEE value means that the implementation provides good results. The calculation for OEE after the implementation of improvements is as follows:

(1) Availability rate after improvement

There is an increase in the availability rate after improvement because implementing improvements reduces machine breakdown time. The data from the availability rate after improvement can be seen in Table 20.

(2) Performance efficiency after improvement

After improvement, performance efficiency increases because the machine is more controlled with reduced damage and TPM or machine checking. Hence, the resulting output is better after improvement. The data from performance efficiency after repair can be seen in Table 21.

(3) Quality rate after improvement

There is a not-too-far increase in the quality rate after improvement because the previous value is good. The data from the quality rate after improvement can be seen in Table 22.

(4) Overall Equipment Effectiveness (OEE)

The calculation results for the average OEE value after improvement have increased from 67.42% to 77.80% (Table 23). This value has increased because each factor has increased, and the improvements that have been implemented affect the OEE value.

Table 20. Availability rate after improvement

Table 21. Performance efficiency after improvement

Table 22. Quality rate after improvement

Table 23. OEE Value after improvement

3.5.2 Six big losses after improvement

(1) Breakdown Losses

The value of breakdown losses has decreased due to reduced damage time with the improvements that have begun to be implemented. The calculation of breakdown losses after the implementation of improvements can be seen in Table 24.

Table 24. Breakdown losses after improvement

(2) Setup and adjustment losses after improvement

The value of setup and adjustment losses has decreased because the loading time value has improved with reduced damage time, even though no overtime is added to have a good loading time. The calculation of setup and adjustment losses after improvement can be seen in Table 25.

Table 25. Setup and adjustment losses after improvement

(3) Reduce speed losses after improvement

The value of reduced speed losses has decreased because the actual production time has become better, the loading time has become better, and the amount of production has increased from before. The calculation of reduced speed losses after improvement can be seen in Table 26.

(4) Idling and minor stoppages after improvement

The value of idling and minor stoppages has decreased because the production amount is close to the target amount, and the loading time is improving. The calculation of idling and minor stoppages after improvement can be seen in Table 27.

Table 26. Reduce speed losses after improvement

Table 27. Idling and minor stoppages after improvement