INTRODUCTION

The paper presents a case study where the Lean Six Sigma (LSS) methodology, based on the DMAIC principle, was applied to improve the repair processes in a metalworking company. The study aims to enhance the repair services of heavy machinery components used in the mining industry.

The application of this methodology in metalworking companies is a novelty. Proper implementation of the Lean Six Sigma approach can help companies increase their productivity, resulting in higher customer satisfaction since machinery downtime will be reduced and production can proceed smoothly.

The study was conducted in a metalworking company that repairs heavy machinery components used in mining, whose name will not be disclosed due to company policy.

The company is facing low productivity in various areas of the repair processes, including Planning, Welding, Machining, Assembly, and Quality Control. To address this issue, the study proposed implementing a methodology to improve these processes, increase productivity, and minimize defects in the repaired components.

After analyzing data from the 2021 period, it was found that the level of defects reached around 50%, with a sigma level ranging between 2.19 and 2.57. The effectiveness was found to be between 70% to 87%, resulting in a productivity range of 0.77 to 0.85.

This study outlines the stages of the DMAIC system and the tools applied, such as SIPOC, Pareto charts, control charts, capability analysis, and T-pairing.

Finally, recommendations will be made for the use of tools that can help improve productivity and meet the needs of customers. It will also be demonstrated that implementing Lean Six Sigma is feasible.

Objective

Improve the productivity of a metalworking company by reducing defective repairs of heavy machinery components used in mining activities.

Specific Objectives

Research Question

How to improve productivity in the repair of heavy machinery components with the application of Lean Six Sigma methodology?

Hypothesis

H_o: Applying the Lean Six Sigma methodology in the repair of heavy machinery components, the productivity of 2022 is equal to the productivity in 2021.

H_a: Applying the Lean Six Sigma methodology in the repair of heavy machinery components, the productivity of 2022 is higher than the productivity in 2021.

METHODOLOGY

A quantitative approach was adopted in this research to collect data on efficiency, effectiveness, and quality performance from a metalworking company workshop. The data collected will be used in a simulation. The research study is non-experimental and cross-sectional since the data was collected at a single point in time. It is also descriptive as it shows a comparison between the current and simulation results to determine the feasibility of using the Lean Six Sigma methodology.

RESULTS

Stage 1: Define

Results from the study showed that the company has a low level of productivity, which could make it challenging to meet customer requirements. The repair processes of category V components, accounting for 45% of the company’s sales, were the least productive. Additionally, the costs of these repairs were valued at over $50,000 (see Table 2).

Based on ^{Arcos and Cerda (2019}), the low level of productivity in the repair process is due to the non-compliance of repair times and a high rate of defects in the repaired components. The entire repair process is illustrated in Figure 2.

To describe how to repair category V components, the SIPOC tool was used. This tool helps outline the steps involved in repairing components in this category. Suppliers and customers provide materials, inputs, and components to be repaired, as described in the tool and Figure 3 (^{Canahua, 2020}; ^{Castillo, 2021}; ^{Corzo & Teccsi, 2019}).

Stage 2: Measure

After identifying the problem, the sigma level corresponding to category V components was measured (^{Alvarez, 2019}; ^{Castillo, 2019}; ^{Carrillo et al., 2021}; ^{Valverde & Reyes, 2021}) as shown in Table 3.

As shown in Table 3, the sigma level obtained in 2021 ranged from 2.19 to 2.57. The target sigma level was 3.4. Moreover, it is observed that the number of non-conforming products ranged from 8 to 15 units, resulting in defects of 52% to 80% in 2021.

Finally, the productivity of repaired parts in category V was measured using the company’s effectiveness and efficiency indicators, which are shown in Table 4.

The productivity data for the year 2021 is presented in Table 4, and it shows that the effectiveness ranges from 70% to 86.67%. The company’s goal was to achieve 90%, but the current levels are falling short of the target. The efficiency of man-hours ranges from 84.54% to 97.34%, which is also below the target. Therefore, the report concludes that the company is not meeting its productivity goals despite adequate use of man-hours.

The productivity results were subjected to a number of tests, and the productivity report for the repaired category V components is shown in Figure 4.

As shown in Figure 4, the productivity data was subjected to a normality test, which yielded a p-value of 0.617, higher than 0.05. A control I chart was used to measure the productivity of the category V components, showing it is under statistical control. However, the variability should be reduced. The productivity capability was measured according to the 0.9 and 1.1 limits and a target of 1. The results showed that the productivity was inherently incapable, with a Cp 0.91 and Cpk −0.64, and the productivity was shifted to the left. Thus, it is recommended that the company stabilize and center productivity results to achieve its targets.

Stage 3: Analyze.

During phase 3, the team analyzed the causes of delayed productivity in the repair of category V components.

Pareto charts were used to detect, minimize, or eradicate the root cause(s). Figure 5 shows the productivity report of the repaired items (^{Guerra, 2019}; ^{Pastor, 2018}; ^{Chamorro, 2017}).

Figure 6 illustrates the application of the 80/20 principle to determine the most frequent occurrences, which were defects in the housings and cracks in weld beads.

The team prioritized the highest incidence, which was “defects in the housings”. Another Pareto diagram was then created to determine the cause of these defects.

The analysis in Figure 6 revealed welding process correction and pores in the filled areas as the most frequent occurrences.

Priority was given to “welding process correction” due to its high incidence.

Finally, another Pareto diagram was created to determine the diagnostic cause, as shown in Figure 7.

After analyzing the data, the team concluded that the low productivity was due to defects in the housings, which were caused by correcting errors in the welding processes. The main cause of the errors was the inadequate welding procedure.

Stage 4: Implementation (Improve)

To address the issues identified, the following proposals were made in order to minimize the causes of low productivity.

Implementation of KANBAN. This system is used to plan and prioritize repair activities based on the level of urgency (red for high, yellow for medium, and green for low). The Kanban board is shown in Figure 8.

Implementing Heinjunka Tool. This tool is used to level the repair activities according to priority levels. A table of goals, subgoals, and limitations was also created to meet customer expectations, i.e., what can be delivered with the resources available to the company. The objective of implementing the Heinjunka board is to minimize delivery time to customers and to improve communication with them (see Figures 9 and 10).

The above table was generated using Heinjunka, based on indicators such as efficiency, effectiveness, nonconforming product, yield, customer delivery time, and productivity.

Implementing a New Welding Procedure. To minimize defects and increase productivity, it is recommended to implement a new welding procedure. The current welding process used by the company under study is SMAW (Shielded Metal Arc Welding), which involves using an electrode to join two metals, also known as coated electrode arc welding process. However, this process has disadvantages such as leaving spatter and porosities that require grinding. An alternative process to consider is the FCAW (Flux Cored Arc Welding) process, which involves welding with gas-shielded flux-cored wire. This process is advantageous as it is tolerant to strong air currents and has a lower cost.

Stage 5: Control Improvement

During this stage, the production area personnel were assigned specific responsibilities.

The planning team was in charge of controlling the production schedule using the Kanban tool. Another team was responsible for implementing Heinjunka, maintaining the equipment, and coordinating with the leaders of the areas involved, such as Machining, Welding, Planning, and Production. Certified welders with extensive experience were assigned to implement the FCAW procedure.

After implementing the improvements in the category V components the results presented in Table 5 were obtained.

As can be seen in the table above, there was an improvement in the sigma level. The table indicates an improvement in the sigma level, ranging from 2.83 to 3.07, bringing us closer to our goal of 3.4. The simulation suggests that reducing non-conforming products by 2 to 6 defects per month will result in an adequate sigma level.

Table 6 shows the possible results of the sigma level.

The data presented in Table 5 indicates that maintaining a productivity level between 0.94630 and 1.02804 requires an average compliance of at least 95% and a monthly efficiency of 97%, achieved through proper resource utilization.

Figure 11 shows the possible productivity results obtained.

Figure 11 shows the improved productivity results obtained, displaying a normal distribution with a p-value of 0.81, which is higher than 0.05 in Graph I, indicating that the monthly productivity indicators are under statistical control. However, decreasing variability is recommended. The capability indicators Cp and Cpk yielded possible results of 1.09 and 0.95 respectively, indicating that the productivity level is inherently capable, but centering is recommended.

After applying the Lean Six Sigma methodology in a simulation, productivity improvements were obtained.

The hypotheses analysis was carried out.

H_o: The productivity of 2022 is equal to the productivity of 2021 when the Lean Six Sigma methodology is applied in repairing heavy machinery components.

H_a: The productivity of 2022 is higher than the productivity of 2021 when the Lean Six Sigma methodology is applied in repairing heavy machinery components.

In Figure 12, a paired t-test was conducted to determine whether the implementation of Lean Six Sigma had improved productivity in the repair of heavy machinery components.

The report indicates that the p-value is 0.000, which is less than 0.05. Therefore, the null hypothesis is rejected, and it is concluded that the application of the LSS methodology is feasible and leads to improved productivity.

DISCUSSION

The improvements that would be obtained by applying the Lean Six Sigma methodology considering the critical business parameters are shown in Table 7.

The implementation of the LSS methodology resulted in remarkable improvements, particularly in the parameters that generate more revenue and reduce costs for the company.

The best results were obtained in the effectiveness parameter, which rose from 79% to 96%, efficiency, which increased from 97.34% to 99.29%, and non-conforming product index, which decreased from 51.68% to 19.47% in comparison to the previous year. Consequently, productivity increased from 0.82978 to 0.98760, enhancing customer perception and the company’s profitability. The company also improved the welding processes by replacing the SMAW welding process with FCAW, which will result in an overall improvement in the company's total production.

CONCLUSIONS