Efficiency has always been a core focus in modern manufacturing, with the pursuit of process optimization being a continuous endeavor. In the early 1920s, Walter A. Shewhart introduced Statistical Process Control (SPC) at Bell Laboratories, pioneering the control chart and the concept of statistical control. SPC principles were later integrated into the management philosophy of Dr. W.E. Deming before World War II.
SPC in manufacturing involves the collection and analysis of key metrics (KPIs) related to the production of parts or products. Its aim is to ensure that the process remains stable, consistent with customer requirements, and free from any variations or variabilities that could potentially lead to the production of defective products. The underlying principle of SPC is to shift manufacturing from an inspection-based model to a predictive and prevention-based model.
The traditional inspection-based model involves building parts and subsequently inspecting them for acceptability. Parts that pass inspection are sold, while those that fail are either reworked or scrapped. However, this approach is not only expensive, with waste destroying up to 20% of every dollar in manufacturing, but is also inefficient, as it may not effectively detect all defects even with 100% inspection.
Over the years, advancements in computational and processing capabilities have expanded the data collection and analysis capabilities of SPC. These tools have allowed manufacturers to identify and address bottlenecks, leading to consistent cost reduction and increased profitability. However, further progress towards achieving 100% efficiency requires a system that proactively detects emerging threats and corrects them in real-time to prevent process failures. This is where artificial intelligence (AI), machine learning (ML), and Industry 4.0 come into play.
By leveraging extensive databases of real-time quality, maintenance, and lab measurements, AI applications can monitor variations across entire factories. They can analyze data from all stations simultaneously, accurately detect emerging threats to the process, and alert key personnel to proactively mitigate these threats. The key elements of a proactive system include real-time process data collection, real-time data analysis capability using AI algorithms, threat detection intelligence through AI and ML, rapid and targeted messaging capability, and a change agent to communicate at the point of operation. Frontline workers play a crucial role as the fifth element in this system, as their engagement is essential for proactivity.
Utilizing AI and ML for the proactive detection of emerging threats in manufacturing processes and taking preemptive actions has several advantages. It helps prioritize threats, ensuring effective resource allocation and attention. Moreover, it drives improvements in key manufacturing KPIs such as Overall Equipment Effectiveness (OEE), First Time Yield (FTY) or First Time Through Quality (FTTQ), and Cost of Poor Quality (COPQ).
OEE is a measure of manufacturing productivity that identifies the percentage of productive manufacturing time. FTY represents production efficiency and quality by measuring the percentage of units without defects or rework. FTTQ measures the percentage of good parts at the beginning of a production run, indicating a well-designed and managed process. COPQ encompasses both direct and indirect costs associated with defects generated by a process. High scrap rates indicate inefficiencies in manufacturing operations, while low scrap rates indicate high efficiency and quality control. Unplanned downtime can also result in significant costs for automobile OEMs, estimated at $50 billion annually.
REVEAL®, an AI product developed by Trumble Inc., monitors the entire manufacturing value stream 24/7 and proactively alerts the workforce to take corrective action before losses occur. The platform automatically mines data and continuously analyzes trends, ensuring manufacturing throughput is protected. REVEAL does not require extensive data mining skills or statistical knowledge and is designed to empower the entire factory workforce. By reducing waste REVEAL® delivers significant savings to manufacturers.
Manufacturers worldwide have implemented various programs and techniques over the past 50 years to reduce costs and improve product quality and customer service. Lean, total productive maintenance (TPM), total quality management (TQM), Six Sigma, and comprehensive production systems have yielded positive results for companies like Toyota, Procter & Gamble, and Danaher, boosting productivity by 6% or more annually.
While these approaches remain relevant, they are no longer sufficient for effective competition in the future. Two critical developments have raised the bar for manufacturers: game-changing digital and Industry 4.0 technologies. When properly deployed, these technologies can raise productivity by more than 10% and offer advantages in flexibility, quality, resource consumption, capex allocation, labor savings, and more.
Digital technologies, such as AI, have the potential to revolutionize production systems by predicting customer demand, optimizing supply chains, enabling self-healing and self-learning production units, and allowing humans to focus on creative tasks and supervision. Leading companies have recognized the benefits of digital integration, with a recent survey showing that those who have successfully integrated digital into their production systems are three times more likely to achieve their system goals.
However, manufacturers also face challenges such as the sustainability imperative, increased competitive pressures, and evolving customer expectations. To address these challenges and unlock value at scale, manufacturers need to pursue a more innovative supply and production system. This approach integrates traditional methodologies like lean, TPM, TQM, and Six Sigma with cutting-edge digital capabilities. It enables manufacturers to accelerate sustainability efforts, optimize processes and workstreams, and transform how they bring products to market.
Many companies fail to fully harness the potential of new digital and Industry 4.0 technologies because they treat them as add-ons rather than integrating them into their existing manufacturing environments. A comprehensive approach is needed, connecting digital technologies with existing methodologies and investments. An end-to-end perspective that aligns digital solutions with lean, TPM, TQM, and Six Sigma investments is crucial for realizing value at scale.
A future-ready supply and production system provides a clear path to autonomous manufacturing. This future state includes digitally enabled shop floors, flexible plant layouts powered by advanced technologies, robust and circular value chains, and adaptive organizations embracing AI-supported problem-solving.
The journey towards this future state varies for each company and plant, depending on maturity levels and ambitions. Cost-benefit analysis, technology requirements, IT infrastructure, and skill development are key considerations. An integrated approach that leverages existing methodologies and incorporates digital technologies is essential for realizing the benefits of the factory of the future.
In conclusion, by embracing AI, machine learning, and Industry 4.0 technologies, manufacturers can proactively control the plant floor, optimize efficiency, and achieve higher productivity. A comprehensive approach that integrates digital capabilities with existing methodologies will help manufacturers unlock the full potential of the factory of the future.
This article was enhanced by the use of AI.