Flow algorithm to enhance operational performance and improve maintenance effectiveness

Manufacturing excellence means the operation works effectively to deliver the output within the desired time, with the highest quality and meeting all specification. The journey towards manufacturing excellence requires smooth operational performance and a strong maintenance strategy to enable it. Problem solving becomes a critical enabler to achieving manufacturing excellence by resolving issues that arise from imperfect operations but problem solving is just half the effort, the other part is the effectiveness of the maintenance practice. Many problem-solving methodologies use very good techniques and approaches to tackle manufacturing problems effectively but do not necessarily look at maintenance within the same effort. The objective of this study was to develop an operational process flow algorithm that would enable user to easily follow to get to a solution and also positively influence the maintenance practice so that both the problem and the maintenance practice are comprehensively improved. The algorithm works by analyzing the problem and narrowing down to a specific problem area of the operation through function modelling analysis and identifying the main parameter of value (MPV) of the problem. Next, the MPV is compared to the current maintenance practice to determine if its performance is sustainable. If not sustainable, then a new method to improve the maintenance is worked out by firstly making the MPV parameter quantifiable. Performance of the MPV is actively tracked by using stability and capability analysis to determine when best to intercept and make the change via a risk assessment matrix. The developed algorithm was then validated on the laser marking operation with both the equipment and operational process performance being validated. The laser diode performance was successfully characterized by analyzing health of diode at different stages of life vs. its measured power output and active monitoring of the laser power and laser current offset performance at the start of each production batch. This allowed the maintenance practice to change from run-to-fail to condition based maintenance (CBM). For the process performance, its maintenance practice improved by successfully enabling optical character recognition (OCR) that allowed 100% active monitoring of all marked products and eliminated manual visual inspection. This was done by building an imaging baseline of characters that are used to compare with each marked product. Any rejects detected are segregated and weeded out within the operation itself. The improvement enabled the overall effective unit per hour (EUPH) to increase by 8% and changed the maintenance from a PM to a CBM. In conclusion, the new flow algorithm was successful in resolving manufacturing problems and improving maintenance practice as a comprehensive overall solution.