Real-time machine tool data collection isn’t just about helping manufacturers improve productivity and profitability, although that’s certainly a promised outcome. It’s also an essential first step toward a data-driven, high-tech manufacturing sector that is globally competitive.
Today’s data-collection and monitoring solutions already help manufacturing operations management to see, analyze and quickly act upon time-sensitive data coming off the shop floor. While these new solutions are more readily available and also less expensive than in the past, they are still installed only at a relatively small portion of factories and shops, with some industry insiders estimating that fewer than 5% of machines are currently being digitally monitored.
Machine tools connected through MTConnect, an open-architecture, royalty-free protocol for machine communications, or through a variety of proprietary protocols available from CNC controls suppliers, can turn a growing stream of data coming off the plant floor into useful information. Key operational metrics such as overall equipment effectiveness (OEE), machine uptime and capacity utilization allow shop managers to quickly view the performance of a particular machine or factory, and adjust as needed.
The coming onslaught of connected Internet of Things (IoT) devices and the Big Data deluge may prompt more manufacturers to seek rock-solid solutions for dealing with extremely large amounts of operational data.
More technology providers, such as Cisco, are moving into the manufacturing space with solutions for handling large amounts of unmanaged data, said David McPhail, CEO of Memex Inc. (Burlington, ON, Canada), developer of manufacturing execution systems (MES). “Within our industry there are 20 million CNC machine tools installed globally, two million in North American alone,” he added, “and three more support assets are typically associated with each CNC, which creates an 80-million-machine universe. Yet over 90% of the world’s CNC assets are not connected because of a plethora of different data protocols.”
As an increasing amount of this data is captured, those within manufacturing see a higher-tech, data-driven manufacturing industry taking shape. Big Data is just one part of the “Third Platform” of computing—a convergence of mobile computing, social media, the cloud and Big Data, he said.
“As this third platform takes hold, plant managers of manufacturing companies are beginning to understand they cannot manage what they cannot measure,” McPhail said. “IoT or the Industrial Internet of Things [IIoT] has some boardrooms buzzing about the possibilities for data-driven manufacturing, so it’s helpful to us, but it’s still a high-flying concept that could take several years to land on earth.”
Helping this concept land is partly depending on that first step—capturing and analyzing data that offers the most value, which isn’t as simple as it seems considering the vast amount of data available in a manufacturing environment.
“Our developers are constantly improving the processing of the immense quantity of events that can be streaming into a database from a shop floor,” said Jody Romanowski, CEO of Cimco Americas LLC (Streamwood, IL), a developer of data-collection software solutions. “When a new system is implemented, it is important to determine if the information will be valuable to you and plan the collection of relevant data carefully.
“It’s easy enough to collect ‘Big Data,’ but make sure you are collecting events from which you can assemble useful information,” she added. “I think some of the Big Data arrays may be filled with information that is not relevant or can’t be processed into useful information.”
Shops today are looking for ways to streamline operations, improve productivity, and reduce cost, noted Mohamed Abuali, CEO of Forcam Inc. (Cincinnati and Friedrichshafen, Germany). “More manufacturers are embracing advanced technologies,” Abuali said. “At large enterprises, there is a necessity for global manufacturing intelligence and benchmarking, where managers can access KPIs and analytics at their fingertips, anytime, anywhere, in any language. At smaller manufacturers and job shops, there is a growing desire to monitor the shop floor, understand part flows, planned versus actual performance, run programs, and access paperless information.”
All manufacturers are trying to enhance the skills of an aging workforce and use technology to drive productivity, he added. “At Forcam, we offer an scalable solution that can address the needs of large and smaller manufacturing enterprises, via a cloud-based or on-premise solution and the needed shop floor management training to assist them to use smart data to make smart decisions.”
The future lies in development of solutions that allow manufacturers to be more proactive, intelligent, and informed, and to support executing timely decisions, Abuali said. Forcam’s platform enables connectivity not only to machines, but to any IT system, such as quality, maintenance, PLM, and tool data, he added.
“MTConnect standardizes links between systems, applications, and entire factories to provide an integrated overall manufacturing system. In addition to MTConnect, Forcam’s technology supports a variety of real-time plug-ins to machine CNCs and PLCs, including manufacturer-specific drivers.
“Forcam is a pioneer of Industry 4.0—the 4th industrial revolution, also called the ‘Industrial Ethernet.’ A major theme of Industry 4.0 is to build a cyber-physical system to map production ‘Big Data’ in real-time, in a virtual space, like a ‘cyber mirror,’” he said. “Imagine your factory modeled in the cloud; this today is a reality. Big data requires advanced technological features such as in-memory technology for real-time data acquisition and processing. With this objective data, wastes are demonstrated crystal clear and you win the largest transparency about the state of your manufacturing. You can analyze the production virtually and optimize it in real-time.”
Many of the disconnected machines on plant floors are older legacy equipment representing a huge investment, and these machines can require more technology than a standard MTConnect adapter to fully implement modern data collection, monitoring and analysis systems.
Reliable costs and quality are achieved by data-driven awareness on the shop floor, said Memex’s McPhail. “Second and third-tier manufacturing suppliers must be able to demonstrate that they can capture, analyze and share their production data with their upstream partners. The Overall Equipment Effectiveness [OEE] metric is now table stakes in the manufacturing business,” he said. “OEE equals availability multiplied by quality multiplied by performance. The ability to demonstrate OEE as well as a process to constantly improve plant efficiency is the basis of coveted trusted-supplier relationships.”
OEE is an ideal, industry-accepted metric, McPhail added. “An OEE of 80% or more is achievable by most manufacturing industry shop floors and the ‘money people’ understand this, yet the majority of shop floors can’t produce this measurement on any kind of consistent basis.
“I wish I could say that clipboard data gathering and spread-sheet analysis was not as prevalent as it was, but that’s the number one ‘app’ out there right now,” McPhail said. “A shocking number of manufacturing CEOs tell me their major challenge is being told one thing about plant operations and seeing something completely different on their income statements. The typical clipboard and spreadsheet approach to capturing OEE, when it is captured at all, delays the time CEOs can get insight as to what is actually happening on the plant floor.”
Taking the MTConnect route to shop-floor monitoring has many advantages. As McPhail noted, MTConnect enables manufacturing equipment to provide data in a single structured XML format rather than an obstructive array of proprietary formats. With uniform data available from both legacy and modern production equipment, sensor packages, and other hardware, software applications can enable more efficient operations, improved production optimization, and increased productivity.
“As transformative as it is, MTConnect still suffers from a basic awareness problem,” McPhail said. “Although machine tool companies like Mazak are pushing MTConnect from its CEO on down, plant managers who may have heard of the standard don’t know exactly what it is or how it can help them.”
Memex’s MERLIN (Manufacturing Execution Real-Time Lean Information Network) MES system is an industrial-strength shop-floor-to-top-floor communications platform that provides efficiency metrics in real time. According to McPhail, MERLIN delivers a 10-50% average productivity increase, and earns 20%-plus profit improvement based on just a 10% increase in OEE. “Payback is achievable in less than four months with Internal Rate of Return [IRR] greater than 300%,” he said. “We connect to any machine, old or new, utilizing native MTConnect, OPC, FOCAS or FANUC I/O link using MERLIN hardware adapters.”
More manufacturers are implementing manufacturing data systems because their customers are demanding that they provide accurate real-time and historical information on part production, said Cimco’s Romanowski. “In addition, shops want information that helps them forecast more accurately, pinpoint production issues, send real-time alerts when production problems occur, better understand their plant performance and provide traceability to the machine level,” she said. “With the correct information and reporting, manufacturing companies can improve interactions with their customers, improve quality, improve efficiency, reduce maintenance and reduce inventory. The technology is constantly evolving to meet these requirements more efficiently.”
Cimco’s manufacturing data collection system, MDC-Max, provides customized reports, graphs, real-time alerts and live screens showing real-time production data. MDC-Max is fully integrated with Cimco’s manufacturing data management database product, Cimco MDM, and its NC file management system, NC-Base. “Since our products are fully integrated, information from our databases can be used in MDC reports,” Romanowski said. “For example, if a part cycle time estimate is stored in the database it can be used for comparison to actual cycle times in MDC-Max reports.”
Among shop-floor monitoring solutions, manufacturing consultant TechSolve Inc. (Cincinnati) offers its Viz Products monitoring solutions as an attractive option for smaller shops, noted Ron Pieper, TechSolve product manager. The company offers multiple layers of adapters in its Viz line.
“There are a lot of people saying ‘Let’s get MTConnect done,’” Pieper said. A large aircraft maker used the MiniViz to start up a monitoring system where none was previously installed. “They have a chunk of machines that haven’t been monitored. They like our concept of MiniViz because it’s a lightweight, tiny application running on an attached PC.”
TechSolve’s ShopViz is a more extensive solution, but both are fully MTConnect-compliant, Pieper added. “The standard is driving in a direction where there’s efforts moving into inspection and quality,” he said. “The more that we can learn about the machine and the metadata, the more it will help us to understand what’s going on with the genealogy of parts and what’s collected. I’m seeing a greater acceptance of the concept of monitoring, from users and from machine tool builders and control suppliers. We’ve got a perfect storm [with Big Data and IoT], what’s going to happen next? There’s going to be a big merge in collecting and interpreting the data.”
A major issue is deciding how much data to collect, and TechSolve positioned MiniViz as a tool to get people started monitoring. “We caution how much to collect. If you try to present everything under sun, you’re going to get knocked over—it’s a fire hose,” Pieper said. “There are an immense amount of variables. You don’t need very much to get a big, big improvement, and you can get into it less expensively.”
Although Big Data casts a big shadow over manufacturing, many see it as more of a future problem. “It still looms in the future. There’s a lot of talk about it but in practice, we’re very much driven by what we need today,” said Jim Finnerty, product manager, ShopFloorConnect, Wintriss Controls (Acton, MA). “It’s important to know what’s coming.”
Shop-floor customers today are looking for integration of software like Wintriss’ ShopFloorConnect data collection and monitoring with existing ERP and MES software, he said. “We tie into their existing software and we have visibility of what’s going on on the shop floor at any one time. We can make the scheduling software come to life.”
With ShopFloorConnect, users get an integrated solution that’s highly customizable. Real-time customizable notifications are the latest tweak to Wintriss’ ShopFloorConnect, Finnerty said, greatly speeding up data delivery. “We’re all about increasing efficiency and reducing downtime,” he said.
One of the biggest issues for shop floor managers is knowing what to measure and what to ignore. Recent research at the National Institute of Standards and Technology (NIST; Gaithersburg, MD) focused on helping manufacturing operations decide what to collect from the shop floor.
NIST’s project is its Smart Manufacturing Operations Planning and Control Program, noted Moneer Helu, mechanical engineer, NIST Systems Integration Division, Engineering Laboratory, who was a Manufacturing Engineering “30 Under 30” award recipient in 2014. The NIST program is developing a product lifecycle test bed built on a cyber-physical infrastructure to enable smart manufacturing research and development, with the goal to help spur US manufacturers’ competitiveness with better factory data-collection methods.
“What can you do with that data? How do you enable innovations, and get some value from that activity? They want to know what types of technologies are needed,” Helu said. “Measurement science is very much the bread and butter of NIST. We want to understand not just whether we have the right data, but if we’re using the right way to collect it.”
NIST is also collaborating with TechSolve to develop case studies in relevant areas of manufacturing, Helu said. The Smart Manufacturing test bed involves networking NIST’s machines both in the Gaithersburg headquarters and NIST’s Boulder, CO, offices.
“We’re functioning basically as a small to medium-sized job shop and building a cyber component with computer-added technologies,” Helu said. These would include advanced manufacturing capabilities such as additive manufacturing, model-based enterprise, machine-to-machine communications, and cloud computing and services, with the aim of collaborating and creating highly customizable products that are faster, cheaper, better, and greener.
Some of the questions that need answering for shop-floor data collection and smart manufacturing include dynamic scheduling and routing and prognostics for predicting manufacturing maintenance issues, Helu said. “Where are those pain points? What are the problems? We’re applying that fundamental research side and focusing on transferring that to industry,” he said.