Lean manufacturing is a set of continuous improvement methodologies that help eliminate waste and maximize efficiency, effectiveness and productivity. The core value at the heart of every lean manufacturing strategy is to maximize value addition and minimize or remove non-value activities.
The lean manufacturing principles that are widely practiced today evolved from a management philosophy instituted by the Toyota Production System (TPS). Its basic tenets focus on ways to eliminate possible waste within a manufacturing system in the automotive space but its agnostic relevance earned acceptance in a wide variety of industries.
Lean manufacturing’s business methodology is driven by the principle of continuously eliminating the seven deadly wastes in any manufacturing process.
Eliminating waste is the core goal of lean manufacturing. Waste reduces the intrinsic value of any process and eventually renders the product being produced less competitive for the manufacturer and less valuable for the consumer.
Overall Equipment Effectiveness (OEE) is one aspect of manufacturing that has consistently challenged industries, especially in terms of utilization and efficiency. The quality (scrap) aspect of OEE is known to have more limitations when it comes to improvement. It has constraints like complex part design, machinery process capability, non-availability of advanced metrology tools, and availability of skilled labor. On the other hand, machine utilization and efficiency – the scope of improvement largely rests on internal and controllable factors. They are also considered the low hanging fruits of OEE. IoT powered devices can help gather information like machine downtime due to waiting for setup even after SMED implementation, non-availability of materials, and operators. Data from auxiliaries like collecting bin and scrap bin can also be easily collected continuously and monitored in real-time with the IoT sensors and data analytics. While SMED can handle Quicker setups, machines waiting for setup are purely data driven. Hence, these platforms play a vital role in lean implementation for sustained improvement.
Usually when it comes to lean manufacturing, process improvements are short lived because of the lack of continuous data for monitoring and sustaining the improvement. Capturing of big data allows the ML platform to run applied data analytics for various scenarios and thereby providing insights for lost production hours along with units. In other words, these applied analytics provides manufacturers with opportunity hours for more production (opening up existing / new additional capacity) along with increased efficiency for peak performance.
It can help optimize production runs with an out-of-the-box recommendation engine that identifies the key settings that contribute to peak performance. It can also provide recommendations to replicate those runs more consistently. Real time alerts offer the ability to act quickly, avoid problems and continually realize cost-saving improvements. It can assist factory personnel to proactively address issues and resolve problems quickly with alerts that reduce quality failures.
Machine Learning can integrate and structure data from multiple sources including machines like ERP, Standalone Quality Systems, CMM tools, and MES systems. Recent AI Models based on applied analytics are trained and validated enabling teams to solve problems and make quicker informed and data driven decisions. This tremendously empowers the CI team and Operations to constantly evolve and improve the Lean Implementation methodologies. ML and Applied Analytics play a critical role in the implementation of effective lean manufacturing.
Step 1: Identification of areas for improvement
Step 2: Lean assessment
Step 3: Defining the scope
Step 4: Selection of lean tools or techniques
Step 5: Training and implementation
Step 6: Kaizen events
Step 7: Pre and post implementation analysis
Step 8: Best practices and lean metrics for sustenance
Step 9: Move to next project