- Product Recalls: IoT-enabled Traceability – A priority for every company across industries
- Challenges in old style Traceability: Expensive and error-prone Manual approach
- New Age traceability: Real Time data, Data Mining and Advanced Machine Learning techniques to track and trace every component of the final product
In a move that surprised many, automobile majors like Ford and GM announced major layoffs to the tune of thousands coupled with plant shutdowns across many major states in USA. Reasons were quoted like to manage supply and demand or cost overruns. On the other side of the world, Samsung, an electronics major, had to recall around 2 million units of the Samsung Galaxy Note, one of its marquee products. As per a noted securities company, the decision to ditch the Note 7 would cost Samsung $9.5 billion in lost sales and put a $5.1 billion dent in profit between Oct 2016-of 2017. These two incidents although seem as different as chalk and cheese on the surface, point to the same root cause: lack of traceability.
Product recalls are not just inconvenient & frustrating for the customers but are also a reputational & financial nightmare for the manufacturer. They are not only limited to Automobile or telecom but have touched almost all industries you can think of -like food, medicines, and toys too. Especially for manufacturing, recalls are particularly catastrophic due to the complex mesh of steps & ecosystem the product goes through before reaching the end user.
In the event of a product recall, the manufacturer should be able to trace back all the way to root cause of the problem to prevent further damage and take corrective action in order to prevent it happening in future. In supply chain traceability importance, “Traceability can quickly identify a problem and find where it started. In that case, you still might have to destroy products, but you won’t have to do a recall,” says Simon Ellis, Practice Director for supply chain strategies at Framingham, Mass.-based IDC Manufacturing Insights.
In the age of hyper-competitiveness, rising customer demands and stringent legislation, the manufacturer of today looks at traceability is a long-term strategy. ISO 8402 defines traceability as ‘the ability to trace the history, application or location of an entity by means of recorded identifications’. For discrete manufacturers, traceability means tracking & tracing every component of your final product- from its suppliers to the final customers. Whereas for process manufacturers, traceability means the ability to trace each ingredient of the final product through a batch genealogy. Traceability enables manufacturers to gain visibility in processes to achieve Just-In-Time (JIT) delivery, lean manufacturing, controlled inventory, enhanced quality, and regulatory compliance. Apart from post-event measures like product recall and product liability prevention, traceability is invaluable for preventive measures like Quality & Process improvement, supply chain, after sales service and cost management.
Challenges in old style traceability:
Until a few years ago, many manufacturing organizations used manual approaches towards traceability. In a postmortem, most of the times, key players spent an enormous amount of time and resources to go through piles of data. They found the needed data by sifting through old paper-based records or spreadsheets, old filing cabinets & log books. Over the years, even though a significant amount of process is automated, a lot of manual intervention was still required making the process expensive and error prone. Even for software that collects data electronically, if the analysis is manual, then data analysis for applications like continuous process improvement, root cause analysis, and process excellence also suffers through the same grill. Even for predictive analytics, traditionally, predictive models were created by taking a sample of data from a sample of assets and generalizing these for predicting issues on all assets. While this can detect ordinary issues, which otherwise are detected in the quality control process itself, but it fails to detect the rare occurrences (that are not frequent) that cause the massive recalls. Although, highly sensitive generalized models can be created to detect any and all deviations but those then generate a lot of false positive alerts which cause a different series of problems altogether. Thus, without real time data & data analyzing techniques the manufacturer of today is very helpless in event of a recall.
New age traceability:
Thanks to the technology advancement, Internet of Things is making great headways into end-to-end traceability. Bringing Real-time data to the fingertips of the decision makers, IIoT collects data from all machines/tools, along with details of the finished product at each stage (for discrete players) enabling a complete blueprint of the value chain that leads to the tracking of the product as well as a process. Advanced data mining techniques and machine learning take care of the other side of the equation, by rapidly sifting through the gigabytes of the collected data to detect event the smallest anomaly that might trigger off a defect of any sort. Thus, with IoT, by connecting the right system and machines with advanced analytics backed by the power of the cloud, makes ultra efficient and reliable production a reality.
Traceability has come a long way from being associated with doom and gloom for the enterprise- like product recalls and part failures. In the next part of our blog series, we will focus on how IoT-enabled traceability can be the manufacturer’s best friend to help every department optimize itself and take manufacturing to the next level.