May 6, 2011
Predictive Analysis for Quality Management
A few days ago, Industry Week published an article titled Prescriptions for Supply Chain Success. The article focuses on building quality early in the process, instead of relying on "back-end testing". It also focuses on the importance of having visibility into supply chain quality.
Our customers have invested in our real-time SPC software to accomplish this objective. Having access to the predictive analysis tools that detect potential quality issues during the production process is critical to ensuring final product quality. Relying solely on post-production inspection is costing manufacturers an enormous amount of time and money on having to scrap or rework defective products.
The article states: "As an industry, it is important to focus quality efforts on those points in the process that are most critical in determining the final product quality. The current paradigm of inspecting for quality at the end of the process is costly and not foolproof. New ways have to be developed to ensure that quality is being built early on in the process versus back-end testing." This article was written by two senior directors at the management consulting firm, Accenture, so it somewhat surprises me that Statistical Process Control wasn't mentioned here. Using real-time SPC software, manufacturers don't have to develop anything new to accomplish this objective-- they simply need to implement the software to take advantage of real-time alerts and analysis of production data.
The article also states that manufacturers must have a "thorough knowledge of their quality processes, a good means of storing and analyzing the relevant production data, and a thorough corrective action plan that gets at root cause and remediation in a timely manner." Our customers have the systems in place to adhere to these best practices. What jumps out the most to me here is the issue of time, which is critical on the production floor. The sooner a potential issue can be identified and addressed, the less costly it is to the business, if it all. Collecting quality data during production and having the actionable intelligence to prevent quality issues is becoming more and more critical-- not just from a cost perspective, but also in terms of avoiding recalls. The article focuses on FDA recalls in the pharmaceutical industry which are becoming increasingly common.
One of the key challenges, the article says, is a lack of integration and disparate IT systems. "It is often difficult for many companies to see impending early warning signs of potential quality and compliance breaches." Supply chain visibility is key. Not having a complete understanding of the quality of incoming materials that are to be used in the finished product can be extremely damaging to the business. Our ProFicient on Demand solution provides this depth of visibility, while also addressing the challenge of integration and disparate IT systems. If both the supplier and the manufacturer are using ProFicient on Demand to view real-time production data, then they can collaborate to ensure that the materials meet specifications. There is no need for complex IT integration. In fact, little (if any) IT involvement is required at all. Using the SaaS (Software as a Service) model, users simply need a reliable Internet connection to share quality data.
"Ideally, you want to be able to predict the potential deviations and make the small corrections before they result in the more visible recalls and warning letters. Predictive analytics allows this to take place." Well said.