Recently, Adam Bluemner, Project Specialist Manager at FindAccountingSoftware.com had a discussion with Milosz Majta, DELMIA Apriso Quality & Maintenance Product Manager at Dassault Systèmes. The theme for their discussion was the benefits of tying Computerized Maintenance Management Systems (CMMS) with quality and manufacturing intelligence to create a predictive intelligence scenario whereby machines can be kept running at or closer to optimal performance as a way to increase quality and reduce downtime. The best way to do so is to approach equipment maintenance management on more of a holistic perspective, as an integral part of manufacturing operations management. Ideally, the best results occur when your manufacturing operations systems can seamlessly connects to your ERP systems so all reporting and tracking requirements can be met, adjusted and improved over time.
Adam Bluemner provided the questions, and Milosz Majta provided the answers, as listed below.
Questions & Answers:
What is leading the pronounced increase in activity and investments in manufacturing software over the past several years?
Milosz: A massive “digitization” is occurring across manufacturing, which began with the product design departments, as evidenced by the profound advances now possible within CAD, CAM and Product Lifecycle Management applications. This growing digital world of design is now making its way down to the shop floor – bringing with it a host of new operations management systems to convert all these digital designs into real products. The recent price drop of 3D printers is just one example of how production processes are being significantly impacted by this transformation.
Why should manufacturers care about how well integrated their CMMS and QMS software systems are?
Milosz: Given all the investment in new applications and IT systems to better integrate shop floor operations with product design, it makes sense that each of the other activities done on the shop floor, such as inventory management, quality monitoring and improvement and equipment maintenance must now keep up, or else those vendor applications will soon go the way of the abacus. Part of this transition to a digital world means interconnectivity is a given. In a digital world, no machine, application or employee can reasonably function in a “silo”. Instead, they must work seamlessly together, so as to enable adaptive operations that can change quickly to meet new opportunities, or, to quickly stop out-of-spec production to minimize the potential cost of recalls or poor quality. An example here would be an SPC alert automatically triggering a calibration order based on the process trending out-of-spec.
How does this new level of integration between shop floor applications such as CMMS and QMS impact the process of managing data originating from ERP, CRM or other business systems responsible for financial or transactional records?
Milosz: As they say, you can’t measure what you can’t track. ERP plays an important role in being the system of record. This means that every transaction must be recorded and preserved in this system to ensure corporate records are reflected accurately and with appropriate transparency to meet audit, regulatory and investor requirements. Shop floor systems, however, are completely different in that they must operate with the highest speed and effectiveness in an environment that is often running 24/7. Given this level of stress on these IT systems, an appropriate architecture must be deployed that simultaneously ensures immediate integration to operations processes, an ability to quickly change these processes, as well as an ability to effectively extract reporting data from the activities to then provide intelligence to the management team for continuous process improvement. Batch-style data uploading is an effective approach to ensuring the systems of record are in alignment with the systems of operations. This alignment can be optimized and best managed when each of your operations systems are seamlessly orchestrated – such as run from a single platform or foundation. This type of layout implies a common data model whereby records and programming logic is shared across functions – as well as different locations – such that all of your manufacturing operations can be run seamlessly as if operating on a single, global plant floor.
Would you agree with the following statement? “Equipment maintenance and quality management are similar in that they become more effective as work within each management discipline transfers from being reactive to proactive.”
Milosz: We always want to be proactive vs. reactive. No one wants to find out that a machine required lubrication after it stopped. In fact, many maintenance organizations are measured by the ratio between reactive and proactive maintenance. Likewise, you don’t want to find out about a quality issue after you have already shipped 100,000 unites to your most important customer! How do you become more proactive? Well, often that is accomplished with intelligence that can be captured while a machine is running. Operators have enough knowledge about their machines to know if a particular performance specification starts to trend out-of-bounds of its normal operating range. Such activity could be indicative of a future issue. What if this intelligence could be instantly captured, processed and then used to alert these operators whenever any such “indicative” behavior begins? That could go a long way towards increasing equipment uptime, or improving quality. This type of scenario is now possible. And, as more systems and equipment is connected together, the more powerful and beneficial such advance notifications can become.
How do you achieve a reliable level of predictive intelligence to ensure that machines can be kept running close to optimal performance as a way to increase quality? What’s the most important data to monitor?
Milosz: There isn’t really one answer to this question. Predictive intelligence is based on building the right models from historical data and predicting future performance, so the more comprehensive and relevant data you are able to use to build the model the more reliable the model will be. It also depends on what you are making and what the process is that is involved. But, what I can tell you is that whatever your process is, you can always get better. And, after a few years of honing your craft, understanding the intelligence that is available and actively using it for performance improvement, you can then turn this knowledge into a significant competitive advantage – an advantage that can not be easily copied or replicated without going through the same learning curve you went through. IN the end, your competitors might never catch up, provided you continue to move forward with your performance improvement journey!
Parts of this interview were reprinted from FindAccounting Software.