Digital Supply Chains: Part Four – Quality, Product Info, Item Onboarding

Capturing Quality Measurements and Taking Corrective Action

A Saudi Arabian manufacturer of plastic products takes measurements of temperature and pictures of the color of the product during production in their plant. These were being recorded on a paper form, which was then signed by the supervisor, taken back to the SAP workstation, and entered into the system. Their SAP system has logic that determines if the batch is within acceptable variation. If not acceptable, they had to dispose of (i.e. recycle and reprocess) the whole batch. Because of the time lag, the entire batch was usually already completed by the time that decision was made. So, they created a workflow-based mobile app that allows them to collect the data, get an answer, and make the decision right there on the production line about whether to continue with this batch or abandon it and start another. Now issues are discovered earlier and a smaller portion of the batch is abandoned, using up fewer materials, energy, and time.

In the next phase, an IoT device will automatically take the temperature and pictures (a small camera takes both infrared and visible spectrum pictures for checking color). The workflow platform will collect and feed that data into SAP and if it is outside the limits, will automatically notify the right personnel to stop the batch via SAP QM (Quality Management) Notification. The workflow platform will notify other interested stakeholders, such as the plant manager. This is expected to cut waste even further.

Automating Complete Quality Management Process in Packaging Manufacturing

A major manufacturer of packaging products makes a large variety of cans and containers made of rigid metal and/or plastic, such as paint cans, ammunition boxes, dairy containers, plastic drums, and aerosol sprays, to name a few. Quality is critically important for their clients who are major CPG brand-owners. The manufacturer is continually striving to reduce the number of quality issues, and the time it takes to respond to and resolve those issues. In the past, whenever there was a quality issue, they used an Excel spreadsheet to collect all the information, such as information about the customer, which plant the can was produced in, lot size, the number of defects in the lot, type and severity of issue(s), whether root cause analysis was needed, and so forth. If root cause analysis was required, there would be an 8D report¹ produced. Often there are associated photos of the issues included. There is also a calculation of the cost associated with the problem. These various files were stored together on a shared file server.

The quality manager would gather much of this information from the customer and inform (usually via email) other stakeholders, such as the account manager and the quality engineer. The quality manager would then lead an investigation, summarize their findings, and create the 8D form if needed. Once the problem was resolved, they asked the customer for approval to close the issue.

There were many shortcomings with this approach. Not everyone fully understood or followed the processes, which could vary by product and plant. The information was often incomplete and people weren’t sure what to put in each field in the spreadsheet. There was poor visibility into how long it was taking to respond, to identify the root cause and then to resolve the problem.

Now this whole process has been automated using a workflow platform. Problems can be entered into a standard form by customer service, sales, quality, or any other department fielding the issue from the customer. As soon as a problem is entered, it gets assigned to a quality manager, who reviews it, categorizes it, and enters the severity, at which point the workflow notifies the plant manager, account manager, QA engineer, and other stakeholders.

If root cause analysis is done by the team, they enter various information into forms. The system automatically generates an 8D report from the information entered and sends it to specified reviewers for review. In this first phase, remediation steps are not part of the workflow, but are being added as part of the next phase as this paper is being published. At that point, the workflow will drive remediation procedures, such as returning and replacing defective materials or components, changing the manufacturing process, or driving an ECO (Engineering Change Order) process.

The new approach saves a lot of time and effort, allowing quality personnel to focus on solving the customer’s problem, rather than on administrative tasks. It also makes the process much more consistent and ensures that nothing is slipping through the cracks. Another big benefit comes from analytics providing visibility into the average time to respond, the average number of days a complaint stays open, and other key metrics. Managers can slice and dice the metrics, such as viewing the top complaint categories by plant and by product types, and drill down to understand what is going on. These analytics enable continuous improvement for the firm.

Reducing Item Onboarding Time by 4X

One mail order/ecommerce company was taking on average two months to add an item to their site and catalog. Up to 10 departments were involved in a serial process where each function ends up waiting for information and steps to be completed by other departments. They wanted to have a more clearly defined and automated process, enabling different departments to work in parallel wherever possible, while maintaining the integrity of the end-to-end process.

They created a workflow integrated with their Dynamics AX ERP system for product master data and SharePoint system holding imagery and other non-ERP detailed data. In parallel to creating this workflow, they were implementing a new PIM system. The workflow platform uses middleware to move data between ERP, ecommerce, PIM, and SharePoint systems. The workflow automatically assigns tasks and tracks progress, and ensures that all tasks are on track and completed on time. Using this approach, they have reduced the time required to create an item from two months down to two weeks. While it reduces the manual labor required to onboard an item, the larger benefit has been shaving six weeks off their time-to-revenue when introducing new items and enabling much better ability to keep up with fast changing styles and tastes.