Data-Driven Manufacturing in the Medical Device Sector

At a recent medical device manufacturing conference, it may not have been the headline topic, but data-driven decision-making was repeatedly mentioned as essential to industry success. 

The realization is reinforced by a recent industry survey commissioned by Dassault Systèmes. Among manufacturers surveyed, 20% reported making bad decisions on a daily basis because of data they did not have, did not trust, or did not have access to. By contrast, the same survey found 15 key areas—ranging from quality control to production efficiencies to sales—where 70% or more respondents reported significant business improvements when manufacturing data was considered reliable and accessible.

Clearly, automated systems of data collection, record keeping, reporting, analysis and visualization are becoming essential for medical device manufacturers. But what’s needed to take all this data, extract the right information, and put it in the hands of medical product manufacturers when they need it? Conversations at the conference pointed to three solutions being increasingly used to automate and centralize data collection, reporting and accessibility: enterprise resource planning (ERP), manufacturing execution system (MES), and quality management system (QMS) software.

Integrated ERP, MES and QMS systems create company-wide information networks that enable organizations to make data-driven decisions in real time. By replacing traditional seat-of-the-pants decision-making, tedious and error-prone manual record-keeping, and disconnected information systems, they are empowering medical device companies of all sizes to improve performance across nearly all aspects of their operations.

Let’s look at two real-world examples of how medical product manufacturers are effectively making data-driven decisions to improve their business. We’ll then review seven pain points and how they can be addressed with better manufacturing data. 

The Power of Manufacturing Data: Case Study #1 

SIGN Fracture Care International is a non-profit company that makes orthopedic nails, plates and screws to repair long bone fractures in remote regions of the world. Operational efficiency for this lower-volume manufacturer is important. But most critical to SIGN are the quality and effectiveness of the medical devices that it ships to isolated trauma centers. The execution of this quality requires SIGN to maintain hundreds of gauges that must be calibrated and documented quarterly. 

Prior to adopting an ERP system with built-in QMS and document control functionality, SIGN’s gauge inspections were recorded on paper and stored in paper files. The record-keeping took hours, and accessing the data was tedious, uncertain and time-consuming. Now those records are kept electronically, dozens of team members can instantly access the data, and missed inspections are automatically flagged for execution. All told, SIGN is saving 20 hours each week by centralizing and providing easy access to this inspection data.

The move has also represented a transition away from relying heavily on institutional knowledge. Prior to automation at SIGN, gauge inspection—including what to inspect, when to inspect, and what was inspected—was nearly single-threaded through one individual employee. Now, it’s possible to have data automatically trigger alerts for events or missed events. For instance, data about a missed in-process inspection can trigger the ERP system to alert a quality supervisor, who can then take corrective action. 

Similarly, automation and easier access to data are enabling SIGN to make timely decisions about inventory. A broken leg cannot wait a week for a femur nail. This means knowing not only what quantities are on hand at the factory but also at the field depots. SIGN now keeps track of inventory in the field, forecasts demand based upon historical patterns, builds to demand patterns using up-to-the moment data, and restocks field depots as needed. So, decisions are based on accurate, concrete data rather than general rules of thumb.

SIGN was a good example of a lower volume medical device manufacturer automating an essential process. Now let’s look at high-volume manufacturer driving decision-making with data from multiple disciplines. 

The Power of Manufacturing Data: Case Study #2 

The medical division of a major manufacturing company makes polymer tubes for reagent test kits for bronchial infections. 

The tests are automated and highly dependent on the test tubes’ uniformity and sterilization. Orders for test tubes are received electronically on an as-needed basis. The equipment that produces the tubes is fully automated. It can produce 16 to 64 tubes in one cycle. All the work center material handling is automated. The parts are digitally inspected at four steps in the production process. The raw material are quarantined and inspected prior to production. The production equipment is fully instrumented for production and process monitoring to record variables, such as cycle times, temperatures and pressures. Production takes place in a secure clean room. 

In other words, the medical division has a state-of-the-art production environment. And the list of data-driven decision making enabled by this production work cell is comprehensive and compelling. 

First, since the sales orders arrive electronically, work orders are automatically created, and the number of cavities to run per cycle can be adjusted to meet the size of a particular order.  Second, because the production cycle time is being measured, an update on the progress of the order through the work center is available at any time along with a prediction of the completion time. Any delays are known in advance, and management decisions can be made to compensate for them. 

Third, the visual inspections catch part defects, and automated process measurements detect production variations—all in real time. Together, the two systems alert process engineering of production issues as soon as they happen, as well as route failed products to the scrap holding center. All of the inspection and process data is immediately available in the statistical process control (SPC) module for root-cause analysis.

Again, through production monitoring, the availability of finished goods is constantly updated so informed customer updates can be made. Similarly, raw material consumption is monitored, and purchasing is made fully aware of the timing requirements for additional materials inventory. At the same time, accounting is automatically updated on planned inventory investments. 

The list goes on, but the point remains constant. The availability of manufacturing data leads to decisions based on factual, real-time information that enable the medical division to strengthen its operations at every step of the way. 

Addressing Top Pain Points with Better Manufacturing Data

We’ve seen two examples of how medical product companies are leveraging manufacturing data today to optimize decision-making and processes throughout the organization. Now let’s recap the top seven areas identified in the survey where manufacturers are using data to improve their performance. 

• Quality Control (47%) – An excellent example of data-driven quality control is in process inspections, which are forced by the quality control module. Operators (or automated equipment) record periodic measurements. If a measurement fails to occur, supervisory personnel are notified. This not only catches defects before they become systemic; it also creates data for later use in SPC analysis and customer documentation.
  
• Business Strategy (43%) – A foundational aspect of managing operations using ERP or similar systems is access to actual cost data, which can drive product and customer scorecards. It helps to identify the products that are most profitable, what products require price updates, and which customers contribute most to the business’ overall profitability—strategic information for managing a business.
  
• Customer Service/Support (36%) – A frequent customer question is, “When can I get my order?”. The capable-to-promise feature of an ERP system looks at raw material, machine and labor availability, process and lead times, and competing schedules to give customer service representatives the information they need to provide fact-driven delivery time frames to customers.

• Operator Performance (34%) – Not all operators perform the same task at the same rate, nor do all machines. Automatically tracking cycle times for both operators and machines provides management with runs-best information, which allows them to assign jobs to the people and work centers that perform them most efficiently 

• Planning/Forecasting (33%) – Using sales orders to create work orders and scheduling those work orders to best meet expected delivery dates is a critical strength of manufacturing ERP systems. The same benefits are gained from forecasting, since knowing historical demand allows manufacturers to plan production in advance and ensure inventory is in stock to fulfill expected sales orders.

• Project Management (32%) – In manufacturing, not everything is a production run. Some tasks are projects like building tooling or installing a new machine. Project management systems excel at guiding and tracking these procedures step-by-step to completion. Just as importantly, they identify projects that have fallen behind schedule or exceeded planned cost.

• Order Management (32%) – Multiple orders often compete for inventory and production resources. ERP systems automatically sort through the constraints and identify the best path forward. This includes not only optimizing production schedules but also identifying resources that need to be expedited or expanded to prevent late deliveries.

Conclusion

The arguments for data-driven decision making are clear and compelling, and the means to that end are often just as obvious. However, medical product manufacturers are often derailed by a lack of integrated information systems. Some smaller companies end up thinking the technology is too complex for their business rather than a toolset to centralize record-keeping and support decisions. Meanwhile, larger manufacturers often have too many best-of-breed solutions operating as data silos, preventing the end-to-end organizational view necessary for informed decisions. 

Increasingly, successful medical device companies of all sizes are demonstrating that comprehensive manufacturing information, made readily available, is a winning strategy that combines the best of industry expertise and data-driven decisions to optimize performance across nearly all aspects of their operations.