Integration between Lean Manufacturing and Project Management in Friction Welding Cell Automation

Bruno Cesar Vilas Boas; Sérgio Ricardo do Nascimento

DOI: 10.22167/2675-6528-202602732

Article derived from a Course Conclusion Work (TCC), with content based on the student’s original work and adapted to the editorial format of the E&S Magazine with the support of the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege for textual synthesis and organization.

Summary

This study evaluated the optimization of an industrial automation project through the integration of project management practices, based on the PMBOK® Guide, and Lean Manufacturing tools. The objective was to analyze the integration of these approaches in the automation of a friction welding cell in the automotive industry, focusing on operational and financial gains. The research was conducted in a friction welding cell in Valinhos, SP, which operated manually and was a production bottleneck due to productivity limitations, ergonomic risks, and low standardization. Value Stream Mapping (VSM), Lean tools (5S and SMED), and time studies were applied. The insufficiency of incremental improvements led to the decision to automate the cell with an ABB IRB 6620 robot, Keyence vision system, and feeding conveyor, implemented in 120 days according to PMBOK® Guide practices. The results indicated a 95.7% increase in productivity, a 50% reduction in cycle time, 37.5% in setup time, and 60.9% in operator displacement. Economically, the project showed a payback of 1.5 years. It was concluded that the integration between Lean Manufacturing and project management optimized the performance of the automation project, generating sustainable gains in productivity, safety, and standardization, with potential for replication in similar industrial contexts.

Keywords: Industrial automation; Friction welding; Project management; Lean Manufacturing; Continuous improvement.

1. Introduction

The growing demand for cost reduction, increased production capacity, and compliance with quality standards drives companies in the automotive industrial sector to adopt automation solutions. These solutions are often associated with lean management methodologies, known as Lean Manufacturing. In this context, the integration between industrial automation and Lean Manufacturing has been widely employed as a strategy to eliminate waste, reduce production bottlenecks, and increase productivity in manufacturing cells (Womack and Jones, 2003; Tortorella et al., 2021).

The principles of Lean Manufacturing allow for the identification and elimination of activities that do not add value to the production process, promoting greater efficiency and reducing variability. Tools such as Value Stream Mapping (VSM), 5S, and the Single-Minute Exchange of Die (SMED) system are widely used to identify bottlenecks, reduce unproductive times, and support automation decisions in industrial processes (Buer et al., 2021; Hossain and Purdy, 2024; Rossini et al., 2021). However, the isolated application of these tools is not always sufficient to eliminate structural bottlenecks, which requires integration with formal project management practices to structure decision-making and ensure execution within the planned scope, deadline, and cost.

This gap is discussed by Ballard and Howell (2003), who argue that lean project management demands the reformulation of traditional planning and control practices. The objective is to incorporate the flow logic and variability reduction typical of Lean Manufacturing, reinforcing the need for conceptual integration between the two disciplines. In parallel, the application of good project management practices, as described in the PMBOK Guide PMI (2021), allows for greater control over scope, schedule, cost, and risks, contributing to the delivery of business value. The most recent edition of the PMBOK Guide, from 2025, reinforces the continued relevance of this reference for project management practice.

The synergy between Lean Manufacturing and project management represents a complementary and robust approach for industrial transformation projects (Campos, 2017; Liker, 2004). The combination of these approaches allows not only for the diagnosis and optimization of processes but also for the structured planning and execution of automation solution implementation, mitigating risks and ensuring that project objectives are effectively achieved. The relevance of this integration is particularly evident in complex industrial environments, where process optimization and project management are crucial for competitiveness.

Given this context, the present work is justified by the need to demonstrate how the integration between Lean Manufacturing tools and project management practices can optimize industrial automation projects, overcoming the limitations of the isolated application of each approach. The objective of this work was to analyze the integration between Lean Manufacturing tools and project management practices based on the PMBOK Guide in the automation of a friction welding cell in the automotive sector, covering not only the financial gains of the investment, but also the operational aspects of the implementation – productivity, cycle time, setup time, and operator displacement – and the qualitative management of the risks associated with the automation project.

2. Material and Methods

The present study was characterized as an applied research, with a mixed approach, combining qualitative and quantitative elements. The methodology adopted was that of a single case study, with the particularity of a participant observer, where the author acted directly as a process technician responsible for the diagnosis and implementation of the project. This approach allowed for the systematic recording of operational indicators, complemented by the practical experience of managerial decisions (Coughlan and Coghlan, 2002; Yin, 2018; Creswell and Creswell, 2018).

The research development was structured in four sequential stages. Initially, a detailed diagnosis of the production process was carried out. Next, incremental improvements were applied using Lean Manufacturing tools. The third stage consisted of the strategic decision for cell automation. Finally, the automated system was implemented, integrating project management practices.

The unit of analysis was a friction welding cell located in an automotive industry in the city of Valinhos, SP. Before the intervention, the cell operated manually and was identified as a productive bottleneck. The characteristics that configured it as a bottleneck included low productivity, high operator displacement, repetitive activities and ergonomic risks, in addition to low process standardization.

In the diagnostic stage, the Value Stream Mapping (VSM) of the current state of the cell was performed. This mapping was conducted in a structured five-day event, involving a multidisciplinary team composed of eight members: six engineers from the forging area, an operator representing the three production shifts, and a Lean Manufacturing (OPEX) specialist. Data for the VSM were collected using a stopwatch, Human-Machine Interface (HMI) records, and interviews with operators from the three shifts, aiming to capture the representativeness of operational variations.

The VSM survey allowed the identification of the main process wastes, according to Womack and Jones’s (2003) classification. These wastes included overproduction, waiting, transportation, excessive processing, inventory, motion, and defects, which characterized the cell as a critical point in the production line and justified the need for intervention.

After the diagnosis, Lean Manufacturing tools were applied. The 5S methodology was implemented to promote the organization and standardization of the workstation. The Single-Minute Exchange of Die (SMED) System was used to analyze and reduce setup times, separating internal and external tooling exchange activities, as described by Rossini et al. (2021).

Setup times were measured over four weeks, with three to four measurements per week, considering that each production batch was approximately 1,500 pieces per part number. In parallel, a time study was conducted using a stopwatch and HMI records from the cell. Measurements were taken over two continuous hours with each operator, covering the three production shifts, with a net productive time of 6.5 hours per shift.

This procedure allowed the capture of natural variability in manual feeding time among different operators and at different times of the shift. Periods of breaks for operators’ personal needs were disregarded from the measurements, focusing only on the effectively productive time. The results were compared between shifts to verify the consistency and representativeness of the collected data.

The operator’s displacement was physically measured by surveying the cell layout. To this end, a spaghetti diagram was constructed for the states before and after automation, recording the paths taken in each production cycle. The analysis of these indicators supported the decision to automate the cell as a solution for the definitive elimination of the production bottleneck.

The implemented automation solution consisted of the installation of an ABB IRB 6620 robot, a Keyence computer vision system for part identification and positioning, with an accuracy of 0.5 mm, and an automated feeding conveyor. The project was structured and executed according to the best practices of the PMBOK® Guide PMI (2021), with an implementation deadline of 120 days.

Project management practices included formal scope planning through the Work Breakdown Structure (WBS), schedule control using MS Project, qualitative risk management, change logging, and the stakeholder communication plan. The implementation involved engineering stages, electrical and mechanical installation, factory acceptance tests (FAT), and operator training.

Change management was conducted according to the PMBOK® integrated change control process, with formal registration of all layout adjustments and integrations identified during execution. A formal economic feasibility analysis, conducted by the controlling and process engineering departments, preceded the decision for automation, considering a maximum payback of 24 months as the investment approval criterion.

Regarding ethical considerations, no data involving human beings was collected directly or indirectly, as the study was based on operational records and observation of the production process. Therefore, the work was not submitted to the Research Ethics Committee (CEP), according to article 1, item VII, of Resolution CNS No. 510, of April 07, 2016, which exempts analysis for research that deepens emergent situations from professional practice without identifying subjects. The reliability of the measurements was ensured by the systematic comparison of data collected across the three shifts, by repeating time measurements over multiple weeks, and by excluding periods with unplanned downtimes.

3. Results and Discussion

The optimization of the industrial automation project in the friction welding cell was achieved through an integrated approach that combined the principles of Lean Manufacturing with the project management practices of the PMBOK® Guide. Initially, the cell operated manually, characterizing itself as a production bottleneck due to productivity limitations, ergonomic risks, and low standardization. The research results demonstrate significant gains in productivity, reduction in cycle and setup times, and improvement in operational safety, corroborating the effectiveness of the synergy between the two methodologies in transforming industrial processes.

The initial state Value Stream Mapping (VSM) of the cell revealed a scenario of inefficiency, with a high throughput time and several non-value-adding activities. Excessive operator movement for manual machine feeding, waiting for cycle confirmation, and manual transport of finished parts were observed. The average productivity was 22.8 pieces per hour, with a cycle time of 2.6 minutes per piece, and a standard deviation of 4.4 pieces per hour, indicating considerable variability influenced by human factors.

To mitigate these problems, Lean Manufacturing tools were applied. The 5S methodology was implemented to physically organize the workstation and standardize consumable items, promoting a more efficient environment. Subsequently, the Single-Minute Exchange of Die (SMED) System was applied, resulting in a reduction of setup time from 120 to 75 minutes, which represents a decrease of 37.5%. This improvement was achieved by externalizing preparation activities, such as pre-heating tooling and separating raw parts, as described by Rossini et al. (2021).

Despite the incremental gains provided by Lean tools, the post-Lean VSM analysis demonstrated that the cycle time and production volume per shift were still insufficient to meet the demand of the production line. The cell persisted as a structural bottleneck, evidencing that isolated improvements were not enough to resolve the fundamental process limitation. This finding directed the strategic decision towards the complete automation of the cell, seeking a more robust and definitive solution to the problem.

The decision for automation was based on a comparative analysis between the implementation cost and the expected productivity gain, with an approval criterion of a maximum payback of 24 months. The automation project, which included the installation of an ABB IRB 6620 robot, a Keyence computer vision system, and an automated feeding conveyor, was structured and executed according to the best practices of project management from the PMBOK® Guide PMI (2021), ensuring a systematic and controlled approach to its implementation.

The automation project management followed the knowledge areas of the PMBOK® Guide. The scope was formally defined through the Work Breakdown Structure (WBS), detailing the automation deliverables. The implementation schedule was planned for 120 days, divided into engineering, installation, and testing phases, and monitored using tools such as MS Project. Change management was rigorously applied, recording layout adjustments and integrations to keep the schedule within the approved baseline, according to the integrated change control principles of the PMBOK® Guide.

Additionally, cost management involved estimating the CAPEX for the robot, vision system, and conveyor belt, establishing a cost baseline for the project. Risk management was carried out through qualitative analysis, identifying, assessing the probability and impact of risks, and defining appropriate responses, which were consolidated in a risk matrix. Communication with stakeholders, including operators, supervisors, and suppliers, was formalized in a communication plan, and quality was monitored through dashboards that tracked the rejection rate and Overall Equipment Effectiveness (OEE) after implementation.

The implementation of the automated system resulted in a significant increase in the cell’s average productivity. Before automation, productivity was 22.8 parts per hour, and after automation, it reached 44.9 parts per hour, representing a gain of 97.0%. This improvement was accompanied by a 50% reduction in cycle time, which went from 2.6 to 1.3 minutes per part, due to the elimination of manual loading and unloading activities.

The analysis of productivity by shift demonstrated consistent growth in all work periods. In the first shift, production increased from 160 to 315 pieces, a gain of 96.9%. In the second shift, productivity rose from 170 to 335 pieces, an increase of 97.1%. In the third shift, production climbed from 115 to 225 pieces, an increment of 95.7%. This uniform behavior of gains across shifts indicates that the benefits of automation stemmed mainly from process standardization, reducing the influence of individual variability inherent in manual operation.

The operator’s displacement per cycle was significantly reduced from 11.5 meters to 4.5 meters, a decrease of 60.9%. This optimization eliminated the main movement wastes identified in the VSM, such as the path from the workstation to the welding machine (reduction from 4.5 m to 1.5 m, -66.7%) and from the machine to the finished parts box (reduction from 3.5 m to 1.5 m, -57.1%). The reduction in displacement contributed to the decrease in ergonomic risks and increased the operator’s availability for higher value-added activities, such as quality inspection and conveyor belt supply.

A relevant finding was the behavior of the standard deviation of productivity, which increased from 4.4 to 8.7 pieces per hour after automation, a growth of 97.7%. Although counterintuitive for an automated process, this increase is explained by the variability introduced by unscheduled stops, resulting from the activation of the “full box” sensor. This phenomenon, identified as a critical operational risk, interrupts the robot’s cycle until the operator performs the manual replacement of the finished parts box, representing a primary opportunity for a future Kaizen cycle in the cell, in line with the continuous improvement philosophy of Lean Manufacturing (Womack and Jones, 2003).

Project risk management addressed several fronts. The highest priority operational risk, the cell stoppage due to a full bin, was classified as critical due to the high probability and direct impact on productivity. The short-term response was active acceptance, with monitoring of stoppage frequency, while medium-term mitigation involves the implementation of an automatic bin exchange system. Technical risks, such as the inaccuracy of the Keyence vision system and communication failures between the robot, HMI, and conveyor belt, were mitigated by preventive maintenance plans, stress tests, and technical support contracts with suppliers.

Security risks, related to compliance with NR-12, were addressed with the implementation of physical and light barriers, inductive sensors, and a periodic audit protocol. Operator resistance to automation, a common risk in Lean projects (Hossain and Purdy, 2024), was mitigated through formal training on the new layout and team engagement since the VSM phase. Finally, the strategic risk of gains being restricted to a single cell was addressed by exploring a rollout in other forging cells, aiming to replicate the project’s success.

In an integrated way, the results demonstrate that, although the isolated application of Lean tools has provided relevant improvements in process organization and setup time reduction, it proved insufficient to eliminate the structural productive bottleneck of the cell. Automation, therefore, represented a technically justified solution, and its implementation was facilitated by the structured project management according to the PMBOK® Guide. This reinforces the complementarity between Lean Manufacturing, as a diagnostic tool, and project management, as an execution and control framework.

The economic feasibility of the automation project was a determining factor for its approval. Previously, part of the demand was met by an external partner, whose unit manufacturing costs were significantly higher than those of internal production, in addition to the monthly freight costs for the return of finished items. Automation allowed for the addition of internal capacity of approximately 3,050 pieces per month, of which 1,784 monthly pieces were internalized, keeping the external partner for contingency flexibility, as recommended by the PMBOK® Guide PMI (2021) for critical processes.

The estimated monthly “saving” of R$ 130,336.00 was composed of two main sources. The first, R$ 45,456.00 per month, resulted from the difference in unit manufacturing cost between internal production (R$ 17.35 per piece) and the external partner’s cost (R$ 42.83 per piece), applied to the volume of internalized pieces. The second source, R$ 84,880.00 per month, was the 80% reduction in monthly freight costs previously incurred for returning finished items to the plant, although the reduction was not complete due to other items produced by the external partner.

With an approved CAPEX of R$ 2,020,000.00, the simple payback calculated was approximately 1.3 years, or 15.6 months. When incorporating the tax burden on financial gains, the adjusted payback was 1.5 years, or 18 months. This value was considerably lower than the maximum criterion of 24 months established by the company for approving industrial automation investment projects, positioning the project as one of the highest financial returns in the plant’s portfolio.

The feasibility analysis was developed based on an optimistic projection of 32.4 units per hour. However, the actual productivity results, which reached 44.9 units per hour, surpassed this projection by 38.6%. This superior performance indicates that the project’s effective financial return is even greater than estimated at the time of CAPEX approval, reinforcing the quality of the planning carried out and the robustness of the time study methodology used as a basis for decision-making.

In summary, the integration of Lean Manufacturing tools and the project management practices of the PMBOK® Guide proved to be a highly effective strategy for optimizing industrial processes. The results in the friction stir welding cell showed significant operational gains in productivity, cycle time, setup time, and operator displacement, in addition to a robust financial return, confirming that the combined approach allowed not only for diagnosis and optimization but also for structured planning and execution of automation, overcoming the limitations of each isolated approach and fully meeting the study’s objective.

4. Conclusion

This study analyzed the integration between Lean Manufacturing tools and project management practices, based on the PMBOK® Guide, in the automation of a friction stir welding cell in an automotive industry. It was found that the initial application of Lean tools, such as 5S and SMED, provided incremental improvements in organization and setup time reduction, but was not sufficient to eliminate the structural production bottleneck. The decision for automation was then conducted in a structured manner, resulting in significant operational gains. An increase of 97.0% in average productivity, a reduction of 50% in cycle time, and 60.9% in operator displacement were observed. Financially, the project demonstrated a payback of 1.5 years, exceeding initial expectations and the company’s approval criteria. The integration of these approaches proved to be a highly effective strategy for optimizing the automation project performance, promoting sustainable gains in productivity, safety, and standardization.

Despite the positive results, an increase in the standard deviation of productivity was identified after automation, attributed to unscheduled downtimes resulting from the activation of the “full box” sensor. This finding highlights the importance of proactive risk management and continuous improvement, central characteristics of both approaches. As a future recommendation, the implementation of an automatic box-changing system is advised to eliminate this residual bottleneck. The study, being a unique case, has limitations regarding direct generalization to other industrial contexts, although it allows for analytical generalization to similar operational scenarios. Additionally, data collection over a short to medium-term horizon did not allow for the evaluation of the stability of gains in the long term, nor the system’s behavior in the face of demand variations. For future studies, more comprehensive investigations are suggested regarding the integration of Lean-PMBOK® in other automation initiatives, with extended time horizons, multiple cell samples, and sensitivity analyses of economic parameters.

Bibliographic References

Ballard, G.; Howell, G.A. 2003. Lean project management. Building Research & Information 31(2): 119-133.

Buer, S.V.; Semini, M.; Strandhagen, J.O.; Sgarbossa, F. 2021. The complementary effect of lean manufacturing and digitalisation on operational performance. International Journal of Production Research 59(7): 1976-1992.

Campos, R.C. 2017. Aplicação do Value Stream Mapping na melhoria de processo produtivo: caso de estudo na indústria automóvel. Dissertação de Mestrado Integrado em Engenharia Mecânica. Faculdade de Engenharia da Universidade do Porto, Porto, Portugal.

Coughlan, P.; Coghlan, D. 2002. Action research for operations management. International Journal of Operations & Production Management 22(2): 220-240.

Creswell, J.W.; Creswell, J.D. 2018. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 5ed. SAGE Publications, Thousand Oaks, CA, EUA.

Hossain, M.M.; Purdy, G. 2024. Role of Industry 4.0 in zero-defect manufacturing: a systematic literature review and a conceptual framework for future research directions. Manufacturing Letters 41: 1696-1707.

Liker, J.K. 2004. The Toyota Way: 14 Management Principles from the World’s Greatest Manufacturer. McGraw-Hill, New York, NY, EUA.

Project Management Institute [PMI]. 2021. Um Guia do Conhecimento em Gerenciamento de Projetos (Guia PMBOK®). 7ed. Project Management Institute, Newtown Square, PA, EUA.

Rossini, M.; Costa, F.; Tortorella, G.L.; Valvo, A.; Portioli-Staudacher, A. 2021. Lean production and Industry 4.0 integration: how Lean Automation is emerging in manufacturing industry. International Journal of Production Research 60(21): 6430-6450.

Tortorella, G.L.; Narayanamurthy, G.; Thürer, M. 2021. Identifying pathways to a high-performing lean automation implementation: an empirical study in the manufacturing industry. International Journal of Production Economics 231: 107918.

Womack, J.P.; Jones, D.T. 2003. Lean Thinking: Banish Waste and Create Wealth in Your Corporation. 2ed. Free Press, New York, NY, EUA.

Yin, R.K. 2018. Case Study Research and Applications: Design and Methods. 6ed. SAGE Publications, Thousand Oaks, CA, EUA.

Article originating from the Final Course Project of the Specialization in Project Management of the MBA USP/Esalq

To learn more about the course, click here and access the MBX Academy platform