Reduce the Total Cost of Quality by deploying AI early and often in your manufacturing process
What is the Cost of Quality?
The American Society for Quality (ASQ) suggests that the Cost of Quality (COQ) is usually around 15–20% of sales, often as high as 40% in some organizations. In other words, the quest for higher quality, and often times the failure on that quest, directly impacts margins and your bottom line.
Cost of Quality is usually around 15–20% of sales, often as high as 40% in some organizations
ASQ defines COQ as a methodology that allows an organization to determine the extent to which its resources are used for activities that prevent poor quality, that appraise the quality of the organization’s products or services, and that result from internal and external failures.
Logically, you would think that the COQ is tied to the Cost of Good Quality (COGQ). These are the costs of trying to produce higher quality goods and prevent the lower quality goods from occurring or slipping through the cracks.
COGQ = Preventive Costs + Appraisal Costs
In fact, on average, 60% of COQ are associated with what is known as Cost of Poor Quality (COPQ). These are the internal costs of defects such as scrap and rework or worse yet the external costs associated with quality issues such as warranty claims, repairs, potential legal liability, and reputational risk.
COPQ = Internal Failures+ External Failures
Failure to correct issues early on in the process leads to far greater costs down the line.
Predictive Analytics in the Quality Process
Through advanced analytics and machine learning, steps can be implemented earlier in the process. By focusing on the COGQ, companies can drastically reduce their COPQ and thus the total COQ as a whole.
Cost of Good Quality
Prevention Cost
Prevention costs make up the smallest portion of total quality cost, but as the first line of defense, they offer the largest area for improvement with the help of AI.
Prevention costs are those incurred from activities intended to keep failures to a minimum. Historically, prevention costs are people and process-oriented such as providing training, developing quality procedures, while also developing product specifications and testing guidelines.
At the prevention stage, machine learning can be utilized to predict the resulting product quality. Given historical information such as measurements or attributes of input materials, combined with machine settings or setpoints, predictive analytics can learn patterns in the data to predict the resulting quality output. In other words, AI can help provide a forecast of the quality of your product, enabling you to optimize settings and reduce the probability of a poor quality result.
Enhancing your prevention measures is the greatest opportunity to prevent poor quality products from being produced thus lowering the chance of higher internal and external costs down the line.
Appraisal Cost
Appraisal costs include the inspection and testing of raw materials, work-in-process, and finished goods. In addition, quality audits, sampling, and statistical process control also fall under the umbrella of appraisal costs.
Tests and audits are performed at various stages of the production process in an attempt to catch defective products. Historically, this has been a very manual process prone to errors and misses. As the age of IoT has evolved, and machines are collecting larger amounts of data, AI can be implemented to automate and enhance these steps, helping to increase accuracy while redeploying valuable employee time.
High appraisal costs are often due to the manual aspect of inspections and audits. Given the manual time-consuming nature of the process, inspections and audits are often completed on a sample of data rather than all products. This allows for poor quality products to slip through the cracks leading to higher external costs.
With computer vision, artificial intelligence can be employed to automate the inspection process. No-code solutions allow employees to build models teaching the system to look for defects that are specific to their business and product line, without the need for a data scientist.
Cost of Poor Quality
Internal Cost
Internal cost occurs when quality defects are discovered before they reach the customer. In other words, a defect has occurred, but the appraisal steps in place have caught the defect preventing the product from reaching the customer.
Defect costs are high as they include excessive scrap and product re-work. Paying less attention to prevention and appraisal costs leads to far greater internal errors downstream.
In addition to defects, internal costs also include the cost of downtown due to machine failure in the manufacturing process. Through the use of machine learning, predictive maintenance can be employed to intelligently monitor the health of critical machinery. IoT sensors combined with AI can provide advanced warnings of machine issues. This allows for planned downtime, smarter utilization of preventative maintenance, and extending the useful life of your machines.
External Cost
Finally, External costs are realized when defects are discovered after the customer receives the product. External costs occur following a large breakdown in the quality management process and they lead to much larger costs to the company. Examples can be related to defective products but also due to products that are of poor quality as a whole. The product may not be defective in the sense that it differs from one version to the next but that all products are of inferior overall quality leading to frequent issues resulting in warranty claims, recalls, and even legal or reputational issues.
AI can help companies look beyond the quality controls related to the manufacturing process. With the explosion of IoT-connected devices, companies can include sensors on their final products allowing AI to intelligently monitor devices in the field and extend their useful life.
Predictive maintenance as noted under internal costs can be employed in your own products to provide superior ongoing quality or your customers, and even to enhance a revenue stream for your organization.
Conclusion
Poor prevention quality measures can lead to high internal costs for your organization. Poor appraisal quality measures can lead to high external costs.
Artificial Intelligence in the form of machine learning offers new tools that can be deployed throughout the quality process. By inverting the common cost of quality breakdown, AI can be deployed at the preventive and appraisal stages. Through the use of AI at these early stages, companies can drastically reduce the total cost while increasing overall quality.
Furthermore, AI can be integrated into finished products to help intelligently monitor their ongoing health and operation. This not only helps decrease costs but also offers new revenue opportunities to companies in the form of extended warranties and other value-added services.
Learn more about how Elipsa is making AI more approachable for quality management: www.elipsa.ai