Abstract
In recent years, the manufacturing sector has been pressured with global warming and resource scarcity. This issue has triggered the industry to seek for solutions to improve the sustainability of their production. Based on literature study, the main components of an organisation consists of manpower, machine, money, material and environment. Thus, the fundamentals need to be addressed to improve the production performance. In this study, an adaptive lean and green approach is presented to identify the priority areas that can improve the organisation performance. Backpropagation algorithm is incorporated into the adaptive model to analyse the dynamic performance of the organisation. However, the input of industry expert is required to prioritise the initial input of the main components. This is relatively important as prioritisation of main components defer from sectors. A case study will be illustrated with the adaptive lean and green model. Operation improvements shall be observed through the implementation of the proposed method.
Introduction: Balancing Efficiency and Sustainability
In the modern manufacturing landscape, the pressure to be both efficient (Lean) and sustainable (Green) is no longer optional—it’s a survival requirement. While Lean Manufacturing focuses on waste elimination and cost reduction, Green Manufacturing targets environmental impact reduction and sustainable development. These two paradigms share significant common ground, such as waste reduction and lead time optimization, but implementing them simultaneously remains a complex challenge.
Most existing frameworks for "Lean and Green" (L&G) operations are static, failing to account for the shifting realities of a live factory floor. Organizations often lack the specialized personnel to manage these overlapping priorities dynamically. This paper addresses the core challenge: how to prioritize improvements across the "4M1E" framework—Manpower, Machine, Material, Money, and Environment—when the optimal priority changes over time.
Methodology: An Adaptive L&G Framework
The researchers developed a multi-stage adaptive model designed to bridge the gap between sustainability goals and operational reality. The framework integrates the Analytic Hierarchy Process (AHP) with a backpropagation (BP) algorithm to create a dynamic control loop.
Initially, industry experts use AHP to weight the importance of the 4M1E components based on the specific sector. A composite Lean and Green Index (LGI) is then calculated using specific sub-indicators, such as employee competency, overall equipment efficiency, and carbon footprint.
The "adaptive" component is driven by the BP algorithm. Borrowing from neural network principles, the model compares the actual LGI against the expected target. If a deviation exists, the algorithm automatically adjusts the weights of the 4M1E components for the next time step, allowing the system to "learn" which areas require the most focus to hit the target.
Case Study and Results
The model was validated using a longitudinal case study of a cogeneration (combined heat and power) plant running on natural gas. The dataset included operational data from 2015 to 2018, with a specific 7-month implementation period in 2018.
Key findings include:
- Dynamic Prioritization: The BP algorithm allowed the framework to shift focus toward Manpower and Machine indicators in late 2018 to close performance gaps.
- Target Accuracy: The adaptive approach narrowed the gap between actual performance and the target LGI (0.96) to a final deviation of only 1.3%.
- Continuous Improvement: Every category in the 4M1E framework showed progressive improvement. For instance, by 2017, the plant achieved a "world-class" performance benchmark of 85% in its Machine (MC) indicator.
Conclusion
This study demonstrates that backpropagation can be a powerful tool for industrial management strategy, moving beyond its traditional uses in AI. While the model currently relies on initial human expert input to set baseline weights, it successfully automates the continuous improvement cycle. Future research will focus on increasing model complexity to handle non-linear interrelations between components, such as how machine efficiency directly impacts environmental waste.
Summary updated