Datanomix Blog

6 Reasons We Need Real-Time Production Scoring for the Machine Shop Floor

Written by John Joseph | Oct 4, 2019 7:41:03 PM


Why real-time production scoring is giving shops 
the actionable insight and data they need from factory machines to drive attention, response, and remediation at all levels.

 

I started my career as a mechanical engineer in the computer industry in the late 1980s. At the same time, my wife and I wanted to own a home, as renting had quickly gotten old in and around Worcester, MA. In order to afford a home, we both worked second jobs.

We had no interest in following the Grateful Dead in a VW Bus or chasing around Europe on a Eurail
pass - especially after my dad took me off his payroll on the day of my college graduation. Graduation was Saturday, and my first day of work was Monday.

 

My wife taught school by day and waitressed nights, while I worked part-time at a machine shop on weekends, running a Bridgeport milling machine. These jobs provided some of the best real-world educations that we’ve received.

 

Working at a machine shop was invaluable in gaining the perspective of what the machinist contends with when he gets a print from an engineer. Seeing the world of design from a manufacturing perspective gave me a great appreciation for what it takes to produce a good part. I really enjoyed this experience because it taught me how to collaborate with people within all functions of the company - each with their own set of deliverables required to produce something precisely and well. It goes way beyond a well-dimensioned drawing. And many of those career lessons are ones that I still practice today. I’m a big proponent of working all elements of the development value chain when you are a young engineer. You can’t just observe it – you must participate and deliver something.

 

Fast forward a few decades, and I find myself back in the machining industry that taught me so much about design, materials, machining, and people - but now using data as a proxy for good production in highly automated environments. An industry that was once dominated by professional machinists, who began as apprentices - with the best becoming Master Machinists - has transformed into a massive floor space filled with CNC tools and robots. Workers are now expected to run two to three machines, not just one, and Master Machinists are becoming scarce, but still precious.  However, one thing remains clear: when the machine is cutting, it is making money. When it’s not, it’s not.

I coined the phrase “the digital machinist” because I now see machining more as program execution, using precise machinery that delivers orders-of-magnitude better precision, than anything I worked on back in the day. The retiring population of expert machinists who cut their teeth on old Bridgeports are being replaced by this digital machinist, who comes in better equipped for automation in this era of smartphones.

 

In 2017, we at Datanomix were working to take machine data and turn it into information people could act on to make their businesses run faster and smarter. Even though I have machining experience, one thing is clear to me – when designing, it begins and ends with the customer. One needs to take the time to understand current business processes, equipment, people, materials, and cash flow in order to deliver a product that hunts. To that end, we conducted several focus groups with companies in the area.  These companies became strategic partners in the formation of our product and, ultimately, the company. A couple of simple open-ended questions in a safe setting with willing participants, who wanted to see change and improvement, yielded some great takeaways:

  1. We need answers at the speed of business. We’re expected to turn a quote in 24 hours, machine quickly, and expedite shipping - and yet, when we need to understand what is happening in the production process that could pose a risk or threat to satisfied customers, we’re expected to wait until tomorrow to find out what and why.

  2. Our systems are antiquated. Manufacturing Execution Systems (MES) have not evolved in 20 years, require tremendous operator input to maintain currency, and rarely yield information that is useful – in the moment. In addition, the user interfaces on these systems are inefficient and cumbersome. We kept hearing “my team hates these systems.”

  3. We need tools to feed strategic analysis. The quote-to-cash cycle is not stable or predictable. The variables required to transform raw materials into known good parts and get paid for them is fraught with variability, with no accurate way of connecting product mix to profitability. When these companies quote parts, they often add a discount factor for manufacturing complexity, which increases the price to their customer to ensure a profit. In some cases, a job is no bid because it won’t make money for the company. There is an active corporate mandate to optimize the product portfolio for that company’s machining competency, machine mix, and specific loading of long and short run parts. 

  4. Resources deployed proactively instead of reactively. The most talented resources are not being strategically deployed across the factory to solve the largest burning issues. We saw the production floors buzzing with activity, and yet the best (and most expensive) people were fighting brush fires all day, instead of focusing on higher level issues which would have a larger direct impact on business outcomes. This, by far, was the biggest burning pain point for company leaders.

  5. How do we augment the skills of our labor force in real time? In discrete manufacturing companies, the priorities around what is being made for which customer by machine and operator changes daily, in most cases - or even more frequently. Manufacturers need systems that are dynamically able to support ongoing decision-making across the factory. Due to variability in the labor force, being able to watch the progress of a job and course correct, if needed, is of the utmost importance to production leaders. Simply walking the floor in modern shops is insufficient oversight of the process. 

  6. We can’t wait until 4:00 p.m. to find out we had a bad day. Modern materials can cost hundreds of dollars per pound. Add to that the expense of machining time and labor, and the idea of waiting until the end of production to find out the process didn’t go well is simply unacceptable.

Introducing Datanomix Fusion

 

Fast forward, and we arrive at the introduction of our Fusion software product. With all the input we received from customers about the information and actionable insight they needed, we focused our time, talent, and treasure on developing a manufacturing analytics framework that does a few simple but important things:

  • Derives its data from the factory machines themselves, with no human input required
  • Augments this data with job-specific insights to create an intelligent and meaningful production score called Fusion Factor
  • Presents information in a way that can be learned, consumed, and responded to in seconds, not minutes or hours

Our Fusion Factor production score is the pulse of the factory floor, driving attention, response, and remediation at all levels of the organization.

 

Fusion Factor monitors production processes, learns what normal production looks like, and assigns a letter grade to both overachievement (A) and underachievement (C) so that the precious resources discussed above can proactively see the trajectory of the run and intervene to get it back on track when it’s not. 

 

This sounds simple on the surface, but I can say with confidence that Fusion Factor is a game changer in manufacturing environments where the leadership is heads-up on company metrics and has a growth compass pointing north east.  Equally critical, Fusion Factor helps create a work environment where there is a vested interest in consistent achievement, and where technology is there to help solve problems -not point fingers. I’ll save the org and compensation discussion for another blog, but we are deployed on well over a hundred machines, and the outcome is measurable. 

 

I have worked on a lot of products, learned tremendous lessons, and delivered great results with wonderful people around me - that’s for sure. At Datanomix we are solving a real problem for manufacturers who are now competing in a global marketplace and need data to win. We make the data analytics business simple. We’re arming them with real-time production information about their CNC machinery and enabling them to act proactively to keep their work on course and speed. Fusion is a secret weapon if you use it. Without Fusion, you’re staring at red, yellow, and green lights on a board, hoping you don’t miss the turn.