Welcome to the Smart Manufacturing Operations, Quality and Value Chain management blog. Talk about Smart Manufacturing and Digital Thread in complex discrete industries like Aerospace, Defense, Medical Devices, Electronics, Industrial Equipment. Share stories, opinions, best practices, lessons learned, and links to related interesting articles and stories.
I can start to see the fog lifting in 2018 from over blown value claims and concerns bolstered over the last two years about implementing new information technology in manufacturing. I believe there is good reason to be excited about the potential of implementing new highly efficient practices, but I am hoping to see more clarity and better understanding of true scenarios, value, and effort to leverage applicable technologies that have become practical to implement in recent years. I am especially optimistic on more clarity in the following five areas:
Security and governance concerns will not to be insurmountable
IT will start pushing back against the unmanageable wave of custom apps appearing at the shop floor. It is easy to build an app with the tools available. But what about security? What about architecting connectivity and data sharing? The current wave of apps is similar to the wave of spreadsheets introduced in the ‘70s when the PC first hit the shop floor. We are creating a myriad of data silos and custom apps that might help for some local optimization but do not necessarily advance the global optimization of the enterprise. IT governance and guidance is required.
Standard approaches to bridging OT and IT networks will emerge. Many stakeholders are working on these types of solutions and the necessary security schemes to connect plant floor operational technology (OT) equipment to enterprise IT systems.
Manufacturing Apps will be seen as complementary and not as replacement for MES
We will see a renaissance of MES. As late 2017 polls indicate, MES is at the top of manufacturers list for implementation because the success stories are abundant, and the word gets out. Do you need a Super MES or MES 4.0? Not really, all you need is a good MES for your specific industry.
There will be new UIs coming out and many MES will get a facelift thanks to the responsive and adaptive UI platforms available. Augmented Reality will become part of the UI landscape as AR instructions authoring becomes more practical for specific use cases.
Hype over IoT/IIoT Platforms will be replaced by better understanding of good use cases.
The practical availability of connected sensors and machines coupled with platforms that can collect, visualize, distribute, and analyze this information is a great opportunity for manufacturing. However, it doesn’t mean we throw everything away and start over either. The new connectivity and platforms enable better and quicker data to be leveraged into improved business processes enabled by new cloud services and integrated enterprise systems including ERP, MES and PLM.
Hype about AI/Analytics/Machine Learning will lead to better understanding of the different technologies under this category and their different use cases
There will be a higher demand for people that understand manufacturing data structures and have data science skills
There will be more realistic expectations for the use of unstructured data in manufacturing.
Manufacturers will uncover that the great value of higher connectivity is not in pursuing better intelligence, it is in pursuing betterorchestration and process optimization that leads to reduced cycle time and increased flexibility. A higher level of business intelligence is a great side effect.
Fear of Robots will be replaced by sensible views on robots as helpers at the shop floor
Instead of taking over all manufacturing jobs, robot use will increase for work that is highly repetitive, high precision, and unsafe.
In many instances, robots will be helping the human worker instead of completely replacing the worker.
We will start to see more robots in managerial functions. Helping managers as sidekicks providing the pre-processing of the enormous amount of data that the factory will generate.
Better understanding of the value and use of these technology building blocks in a future Smart Manufacturing architecture will help accelerate their adoption. The initial hype has opened minds to the potential, but the momentum can fade if we don’t quickly follow with practical examples and guidance on how to thread these technologies together for high impact changes.
A few decades back we might have asked “What is lean about Lean Manufacturing? But nowadays everyone knows what Lean Manufacturing means and what the guiding principles are. For Smart Manufacturing, we are just starting to educate on the principles and how to tell if projects and initiatives are making progress towards a future “Smart Manufacturing Enterprise”. I am confident that a few years from now, everyone will know what it means.
Smart Manufacturing is an endeavor to elevate the connectivity and orchestration of manufacturing processes across the entire value chain to new heights. In Smart Manufacturing, the digital information about each part or product, a.k.a. its digital twin, follows along with the physical part or product accumulating more data as it makes its way through production and inspection processes, supply chain processes, assembly and test processes, and eventually handed over to the end customer as a bundled physical and digital product or service.
Today’s manufacturing reality is far from this digitally connected dream so Smart Manufacturing is an endeavor that will take many years. However, industry leaders are positioning now to be the market disruptive innovators and are making strides to be among the first wave of companies ready to join these new smart manufacturing ecosystems as they are established.
What should manufacturers be doing now to get ready? Companies need to create their own roadmap for the journey and start investing in projects and initiatives that get the company closer to a digitally connected real-time enterprise. The following are some guiding principles to keep in mind for your Smart Manufacturing initiatives:
Minimize manual data entry, translation or transformation of information at each process step.
Processes should hand over structured parsable data (like XML or JSON web services) as output to successor processes to facilitate publish/subscribe mechanisms to systems within the company and into the value chain.
Establish distributed nodes of autonomous diagnostic and decision support at the machine, factory, and enterprise levels. These nodes should roll up transaction information for each machine, factory, and each product unit as needed
Allow ubiquitous use of mined information throughout the product value chain including end-to-end value chain visibility for each product line connecting manufacturer to customers and supplier network.
Automate routine tasks and decisions but include people in the process loop wherever needed to handle nonroutine situations, manual adjustments, and complex decisions assisted by the analytical insights.
Implement optimization schemes that leverage acquired data, advanced analytics, and machine learning algorithms to recommend process adjustments at different levels of the enterprise from controls, to operations, to value chain.
Adopt machine-to-machine (M2M), application-to-application (A2A), and business-to-business (B2B) integrations standards that will enable multi-vendor hardware and software plug and play solutions with open integrations platforms to the internet. Learn about organizations including ISO, IEC, NIST, OAGi, MESA, IIC, DMDII, and CESMII which are establishing and promoting standards for data exchanges among assets, production processes, and supply chain.
Develop a collaborative culture and new workforce skills that encourage projects that cross old functional boundaries like engineering, production, automation, and information technology (IT). Manufacturing automation personnel needs to understand IT and vice versa. Workers will need to learn to configure and maintain smarter machines and robots.
Smart Manufacturing is an evolution to new levels of connectivity over the next few years. Revolutionary productivity gains are expected from the resulting new integrated value chain processes. These are exciting times to be in manufacturing. Make sure you are not a spectator on the sidelines watching others make progress because it could be hard to jump in later and catch up if you wait too long to get started.
I encourage everyone to explore more Smart Manufacturing related resources at www.mesa.org.
This article was originally published at AutomationWorld’s blog on automationworld.com on Oct 19, 2017.
Manufacturing Operations Management (MOM) software remains on the top of the list of prioritized manufacturing investments in surveys by both Gartner [4] and LNS Research [2]. Not only is it on the top of the list, LNS found that 68% of manufacturers were viewing the MOM as part of their enterprise systems strategy.
Why are MOM investments critical to manufacturers?
“MES (aka. MOM) applications boost manufacturing reliability by digitally enforcing processes, methods and procedures. Because MES applications also provide genealogy and traceability information required for compliance, they serve as a strategic nucleus for broader manufacturing architectures. MES data is also an important resource for determining discrete product quality. When scoped, implemented, and integrated properly with other processes for product supply, MES applications can support growth goals of the business while removing costs and complexity, driving process optimization, improving product quality, and directly impacting the bottom-line.”
Quote from Hype Cycle for Discrete Manufacturing and PLM – 2016, Gartner [4]
Leading companies are sharing their MOM success stories in surveys, at conferences and press articles, and the word is getting out. An LNS report [5] shows that implementers of MOM systems have improved Total Cost Per unit by 22.5%, Net Profit Margins by 19.4%, and On-Time Delivery by 22%. Improvements for MOM users were generally double the average among all respondents.
Specific improvements listed by manufacturers include:
Common user interface across the enterprise
Elimination of duplication of input
Integrated interdepartmental information
Platform of interdepartmental collaboration and workflow
Closed-loop quality management
A similar survey performed by Gartner and MESA International [1] showed that most manufacturers implementing MOM/MES had achieved significant improvements within 12 months of implementation. Over a third of the surveyed manufacturers saw results within three months.
Because the MOM serves as a critical link in the information value chain
MOM offers immediate benefits that are easy to quantify for the investment payback. However, manufacturers that stop their ROI analysis here risk missing the larger savings and productivity gains that impact strategic business outcomes. For many companies, MOM has served as a critical technology linking manufacturing operations with customer and supply chain processes.
As Gartner found out in their analysis [3], the largest benefits of MOM come from capitalizing on the visibility it provides into manufacturing performance and capabilities across the organization and supplier network. Participants in a 2015 Gartner study on the value of MOM reported that they were investing in MOM to drive continuous improvement, eradicate variability (that is, improve quality and compliance) and reduce costs. However, they also realized many more tangible and intangible benefits from the MOM implementation the following years.
Although these companies pointed out that the MOM couldn't be solely credited for all the performance improvements, they highlighted its pivotal role in creating an information environment that accomplished the following:
Supported integrative and sustainable improvement initiatives by providing continuous performance feedback through real-time information visibility
Provided the stakeholder visibility needed to manage and streamline the end-to-end supply chain
Created a sense of ownership and performance consciousness among operations and supporting functional personnel
Standardized processes across multiple production sites, allowing for resource sharing and flexibility
A strong US economy and middle class needs a strong manufacturing sector so I would not expect any arguing about wanting manufacturing to be great. However, when we add “again” to the end of that phrase, we need to be clear on what we mean by “again”. If we are referring to making manufacturing popular again then I totally agree with the idea, but if we are referring to taking manufacturing back to older practices then it sounds like a bad idea. Manufacturing practices are much better now than they were 20 or 40 years ago and will continue to get even better going forward as we implement more Smart Manufacturing practices.
Can we bring the old manufacturing jobs back? The straight answer is No.
We can’t because the old manufacturing jobs are gone and not just gone to other countries. The incentive of offshoring to pursue lower labor cost has dissipated over the last few years. [1, 3, 4, 5, 6, 7] Some manufacturers have even been reshoring factories over the last few years. However, it seems that most of the reshoring is also behind us. The manufacturing performed in other countries is usually there to make products sold to those markets. The old manufacturing jobs are gone because the manufacturing industry has evolved and the new processes and equipment require fewer personnel with higher skills.
The chart from MIT Technical Review contrasts the loss of manufacturing jobs against the increased manufacturing output as proof that the loss of manufacturing jobs over the past few decades has been related to huge productivity increases. While manufacturing output has been increasing, production employment has been decreasing, and that trend is not changing. We just do not make things like we used to. [1, 2]
Manufacturing has become a high-tech industry and the related jobs have evolved. There is no going back. Even schemes to tax robot labor will not bring those jobs back. Just like we are not going back to printing blueprints or faxing purchase orders, we are not going back to older labor intensive, unsafe, and error prone manufacturing methods.
Let’s face it, some of those old manufacturing jobs were not that great. The pay was okay, but many of those jobs were dirty and unsafe requiring handling of hazardous materials and dangerous equipment. I remember, during my first manufacturing job, meeting several people with missing fingers due to accidents with older unsafe machines.
The fact that manufacturing jobs have changed should be viewed as a positive and the sector is growing in the U.S. thanks to the use of modern technology. The higher level of productivity of smarter factories allows manufacturers to compete against lower labor cost countries that might be using older machines and processes. Smart factories can produce higher quality products quicker through highly flexible and innovative equipment and systems. We should view this manufacturing evolution as a great opportunity to develop a strong highly skilled workforce.
There are plenty of manufacturing jobs going unfulfilled.
Based on a Manufacturing Institute’s 2015 report, [9] if we continue current trends over the next decade nearly 3.5 million jobs will be needed and 2 million would go unfilled due to the skill gap. This is based on a combination of baby boomers retiring and expected growth in demand for high-tech manufactured goods. If the gap becomes that significant, manufacturers will have to look for alternate places to fulfill the gap. How can we keep those jobs in the US? What skills do we need to develop for the required workforce?
The manufacturing shop of the future is more like Iron Man’s workshop and less like the old factory with smokestacks and assembly lines where mechanics turned the same wrench and the same five lug nuts over and over all day.
We need to figure out how to get more people interested and trained in these new manufacturing jobs. We can no longer bring new people into factories and put them to work without training. Should government help manufacturers with the cost burdens related to training workers on highly specialized equipment?
What skills are needed for these manufacturing jobs?
Technical, computer, math, and problem-solving skills top the list of requirements for modern manufacturing jobs which are becoming increasingly more technical in nature due to the advanced equipment and the integration of digital computer-driven processes in the smart factory. [9, 11, 12, 13]
A recent study by Accenture [10] estimated that a U.S. manufacturer is losing on average 11 percent of their annual earnings (EBITDA) or $3000 per existing employee due to the talent shortage. Other reports estimate that it could be as high as $14,000 per employee. These additional costs are related to longer cycle times, higher overtime pay expense, higher downtime, and higher personnel training costs.
Will higher pay attract more people to manufacturing? Four out of five U.S. manufacturing companies surveyed [9] are willing to pay more than current market rates to hire and retain skilled workers to tackle talent shortage. An average manufacturing worker in the U.S. earned $77,506 in 2013 – 20 percent higher than what an average worker earned in other industries. However, while paying higher wages helps to attract talented candidates, it likely isn’t enough to solve the shortage.
What are manufacturers doing about the skills shortage?
Manufacturers realize they need to improve the perception of the industry as being clean, safe and high-tech rather than dirty and dangerous. [14] These perceptions are improving according to a more recent study but mostly among people closer to manufacturing and more aware of the great technology strides in the industry. [15] Programs like Manufacturing Day are helping change the awareness and attitudes towards manufacturing careers. According to the MfgDay.com website, an estimated 225,000 students had their attitudes toward manufacturing improved in 2016 thanks to over 1,600 Mfg Day events. [16]
Manufacturers are developing more internal training programs, recommending external certification programs, and working with local schools and community colleges to develop the training needed. They recognize the need for an integrated training approach that will depend on these partnerships. They are also more open to using some resources on contract for very specialized jobs like maintaining or programming sophisticated production equipment.
What can the federal government do to help manufacturing?
Manufacturing is of interest around the globe and is an important part of government agenda in many countries because it creates high paying jobs, drives technological innovation, and generates more economic activity than any other sector.
The competition to attract manufacturers is a global one and initiatives to promote and advance manufacturing from other governments include Germany’s “Industrie 4”, France’s “Industrie du Futur”, and China’s “Made in China 2025”. The U.S. should not fall behind in manufacturing technology leadership and the federal government has a role to play. The U.S. strategy to maintain and attract manufacturing needs to includes programs to help train the new manufacturing workforce.
In addition to helping with education, research and development programs, the federal government has an important role in maintaining and enhancing the infrastructure required by the manufacturing industry. See the related article on this site with the title “The Role of Government Supporting Smart Manufacturing” [19] for more information on how the federal government supports the manufacturing ecosystem.
What can states do to attract more manufacturers?
A state’s ability to attract and maintain manufacturers not only depends on the availability of resources including skilled labor, a rich infrastructure of universities, educational and research centers, and a strong network of suppliers of critical components like electronics. But also on the cost of doing business in the state which includes wages (which are tied to cost of living), utilities, taxation policies, and state and local regulations. [17]
Don’t rule out expensive areas. Sometimes cost of living can be offset by quality of life when trying to attract talent. Access to a good transportation infrastructure of ports, rail and highways is also important to help goods in from the supply and out to the distribution network. States should strive to develop areas like Silicon Valley in California with an abundance of suppliers in high-tech electronic components and software developers for embedded systems.
Some states are providing additional incentives to attract business including tax breaks for plant construction and expedited regulatory processes. Additional tax incentives can be provided as a proportion of business activity in the state including wages paid and sales tax generated. The states can also help with the cost of workforce training on the new skills required.
The U.S. is producing a lower percentage of college graduates with advanced degrees than Europe and Asia. The industry needs these higher skills as part of the workforce to support smart manufacturing factories. States with higher graduates in these advanced areas could have an edge.
In addition to the traditional Manufacturing Engineering, Industrial Engineering, and Quality Engineering degrees, more manufacturing certificate programs are needed in skills like CAD Drafting, Tool Design, Industrial Practices, Manufacturing Operations Management, and Quality Control Technology.
Industry-endorsed certifications can also be part of the training mix. These are third-party assessments based on standards that reflect the knowledge and skills required to do the work in a modern manufacturing workplace. For example, the American Society for Quality (ASQ) offers 16 different certifications including Quality Technician, Quality Inspector, and Quality Engineer. These do not substitute for, but rather complement, traditional education credentials. [18]
States can also work with industry to promote more “Learn and Earn” programs, more internships, and mentorships to align higher education with industry competency and skill requirements.
In summary, we want US manufacturing to have a great future and we want to continue the growth trends of recent years. However, growth will be hugely handicapped without development of the skilled workforce required for future smart manufacturing. Government, industry, and academia need to continue working together on solutions to really make US manufacturing’s future a great future.
Government’s role goes beyond helping with education of the new workforce. It also requires investment in infrastructure including smart grid, internet, transportation, communication standards, and research and development institutes to keep America competitive in this new global manufacturing landscape.
References
[1] “What it takes to bring back a manufacturing job”, S. Ben-Achour, Marketplace.org, 2016
Manufacturing is an important part of the government agenda in many countries because it creates high paying jobs, drives technological innovation, and generates more economic activity than any other sector. The competition for a strong manufacturing industry is a global one and initiatives to promote and advance manufacturing from other governments include Germany’s “Industrie 4.0”, France’s “Industrie du Futur”, and China’s “Made in China 2025”. The U.S. should not fall behind in manufacturing technology leadership and the federal government has a role to play.
The U.S. strategy to stimulate, maintain and attract manufacturers has many dimensions including promoting research and development, improving the business tax code, training the manufacturing workforce, and establishing favorable trade policies that open global markets and cut down trade barriers. The environment and proposition for establishing manufacturing plants in the U.S. needs to be competitive with other areas in the world. The country should maintain the tradition of a strong manufacturing industry—built upon a foundation of hard work, determination, and innovation.
Research and Development - The government plays a role in research and development of new manufacturing technologies
One of the ways that government stimulates manufacturing is through research and development (R&D) programs where academia, industry and government collaborate to move the entire industry forward and share the risk and rewards of innovative technology research. R&D activity is vital because economies that have consistent levels of innovation, also tend to have high levels of economic growth [3]. The business sector accounts for most of R&D investment, but the public sector is responsible for approximately 40% of R&D investment in the U.S. [4]
The U.S. federal government’s “Manufacturing USA” initiative has created a network of over a dozen manufacturing innovation institutes around the country including the following:
America Makes - America Makes is a national accelerator and the nation's leading collaborative partner for technology research, discovery, creation, and innovation in additive manufacturing and 3D printing.
ARM (Advanced Robotics Manufacturing) - The ARM Institute’s mission is to create and then deploy robotic technology by integrating the diverse collection of industry practices and institutional knowledge across many disciplines – sensor technologies, end-effector development, software and artificial intelligence, materials science, human and machine behavior modeling, and quality assurance – to realize the promises of a robust manufacturing innovation ecosystem.
CESMII (Clean Energy Smart Manufacturing Innovation Institute) - Smart Manufacturing works to spur advances in smart sensors and digital process controls that can radically improve the efficiency of U.S. advanced manufacturing.
DMDII (The Digital Manufacturing and Design Innovation Institute) - DMDII encourages factories across the United States to deploy digital manufacturing and design technologies, so those factories can become more efficient and cost-competitive.
IACMI (The Institute for Advanced Composites Manufacturing Innovation) - IACMI is committed to accelerating development and adoption of cutting-edge manufacturing technologies for low-cost, energy-efficient manufacturing of advanced polymer composites for vehicles, wind turbines, and compressed gas storage.
Transportation Infrastructure - The government supports nation-wide infrastructure
In addition to helping with education, research and development programs, the federal government has an important role in maintaining and enhancing the infrastructure for utilities and transportation required by the manufacturing industry.
The federal government, working with state and local government, has played a central role in the development of our nation’s transportation infrastructure. In the 19th and early 20th century, the federal government invested in roads, locks, inland waterways, and harbors, and that investment paid off allowing the U.S. to grow into the strongest economy in the world.
In 2002, U.S. freight carriers moved over 19 billion tons of freight valued at more than $13 trillion, and traveled over 4.4 trillion ton-miles over our transportation network. The U.S. Department of Transportation (DoT) estimates that by 2035, the volume of freight shipped on the U.S. intermodal transportation system will increase to 33.7 billion metric tons, worth more than $38 trillion—an increase of more than 48 percent. [1]
Advances in logistics have turned the nation’s roadways into virtual warehouses thanks to “just in time delivery” but the growth in congestion on the nation’s roadways threatens these efficiency gains.
The nation’s rail network has also experienced significant freight growth, and the U.S. DoT estimates that demand for rail freight tonnage will increase by 88 percent by 2035.
Ports and waterways continue to serve as vital links to trade. The U.S. is the world’s largest trading nation, accounting for nearly 20 percent of the world’s total ocean borne trade. 95 percent of all U.S. foreign trade tonnage is shipped by sea. Over the next 20 years, the DoT expects U.S. foreign ocean borne trade to double.
Air transportation is another key part of the transportation system. The federal government has long been involved in promoting and developing the nation’s air service and infrastructure. This role continues today with the Federal Aviation Administration’s (FAA) operation of the air traffic control system and oversight of safety of airlines, pilots and airplanes. The FAA also provides grants to airports to develop their infrastructure through the Airport Improvement program.
Many more manufacturers are dependent on shipment services via airplane—not only from suppliers but also for delivery of products to customers. Airport congestion causes billions of hours of travel delay that result in additional billions of gallons of fuel being used while shippers, travelers, and commuters are stranded in traffic and not moving. This congestion is also increasing logistics costs. Transportation accounts for 60% of the total logistics costs for manufacturers.
Many aspects of the nation’s transportation network are currently operating at or near capacity. With future trade volumes expected to more than double across all modes, new strategies are needed to fund and allocate the resources required to further develop these systems and meet the needs. Insufficient investment, and the resulting inefficiency in the roadway system, is having a significant impact on the nation’s economy.
Other countries including Brazil, China, India, as well as many European countries, are investing in their transportation infrastructure. The U. S. government could provide more visibility on these issues to the public and explain the importance of these types of investments to the future of the manufacturing industry.
Energy and Utilities Infrastructure – The government facilitates the new Smart Grid
Between the late 19th and early 20th century advances in infrastructure, standardization and mass production methods fueled the second industrial revolution. Infrastructure advances included large growth in railroad, gas, water, electricity, telegraph, and telephone networks. Government stepped in and established standards for a range of services including railroad gauges, electricity voltages, layout of typewriter keyboards, and rules of the road for automobiles. This period was marked by large economies of scales created by the new infrastructure and a corresponding reduction in setup and indirect cost to manufacturers.
The U.S. electric grid turned out an engineering marvel with more than 9,200 electric generating units having more than 1 million megawatts of generating capacity connected to more than 600,000 miles of transmission lines. The electric grid is more than a network of electrical plants and wires, it is an ecosystem of asset owners, manufacturers, service providers, and government officials at federal, state, and local levels, all working together to run one of the most reliable electrical grids in the world.
The electric infrastructure is now aging and it is being pushed to do more than it was originally designed to do. The grid needs to be modernized making it “smarter” and more resilient using innovative technologies, equipment, and controls that communicate and work together to deliver electricity more reliably and efficiently. An improved grid can greatly reduce the frequency and duration of power outages, reduce storm impacts, and restore service faster when outages occur.
The “smart grid” is a proposed upgrade to the electrical infrastructure with smarter sensors, controls, digital meters, batteries, and systems that can improve: storage of electricity when we have excess, re-routing of resources based on demand, handling of faults, and consumer control enabled by APIs for business and home equipment. A new smart grid can integrate consumers and suppliers, provide improved security, reduced peak loads, increased integration of renewables, and lower operational costs.
Communications Infrastructure - The government helps to establish rules and guidelines for integrating the ecosystem
It might be counterintuitive because standards are difficult to develop and take time to adopt in industry, but standards support innovation and growth in many ways:
Standards permit the sharing of investment and risks related to implementation of the standard across companies in an industry
Standards help the exploitation of innovative ideas that depends on communication and integration standards as a framework for collaboration in the manufacturing ecosystem
Standards allow innovation by combination of techniques that involve different standards across different domains.
Standards avoid redundant efforts for many companies of “re-inventing the wheel” when a standard practice already exists.
Standards create a more competitive environment by allowing smaller product and service companies to participate with less cost and risk in a more open multi-vendor manufacturing ecosystem.
Throughout the 19th century the American, British, and French militaries were sponsors and advocates of interchangeability and standardization. The standards developed had a positive impact on industry for many years but eventually became too many and a burden on defense procurement processes. In the U.S., defense standards including MIL-SPEC and MIL-STD had grown to over 30,000 in 1990. In 1994, the DoD decided to reform the practices and move from government developed standards to commercial standards developed by non-military standards groups.
Examples of organizations working on standards are ISO (an International Federation of the National Standardizing Associations), and the IEEE (Institute of Electrical and Electronics Engineers). These are technical professional associations with membership composed of engineers, scientists, and professionals from industry, academia and government working together to develop and evolve standards for specific applications and industries. Government now endorses and promotes these standards and uses them in their procurement processes.
Government was instrumental in the development of the Internet which evolved from early government networks like ARPANET and CSNET. In 1974, the IEEE published the TCP/IP standard which became the foundation for networking for the Internet. However, the foundational model for that standard was developed by DARPA (Defense Advanced Research Projects Agency) in the 1960s. In 1982, the DoD declared TCP/IP the standard for all military computer networking. Afterwards IBM, AT&T and DEC adopted TCP/IP despite having competing protocols. The Internet has been embraced as a platform for eCommerce applications and data exchange with suppliers via integration standards like EDI (Electronic Data Interchange) and OAGIS (Open Applications Group Integration Specification).
The biggest use of the Internet is yet to come with the Internet-of Things (IoT) and the convergence of digital and physical worlds in Smart Manufacturing. Gartner estimates that connected pieces of equipment in manufacturing will rise 30% every year to over 500 million by 2020. [7] It would be impossible to be looking forward to the IIoT (Industrial Internet of Things) without standards like TCP/IP, OPC-UA, MTConnect, EDI, and OAGIS helping us connect equipment and systems from the manufacturing factory into the supply chains.
The work of the manufacturing institutes listed above includes evolving, developing, and promoting the communications standards needed for future digital and Smart Manufacturing ecosystems through documented integration practices and test bed results. After all, the digital highway needs rules of the road too. A government agency helping with standards is the National Institute of Standards and Technology (NIST). The NIST paper, “Current Standards Landscape for Smart Manufacturing Systems”, [5] is a great resource for learning more about the evolving role of standards in manufacturing. NIST has also published an important Cybersecurity Framework [6] which many organizations are using as a reference model for reviewing security in manufacturing systems.
Organizations like NIST have a role to develop the framework of infrastructure, models and standards for information management of the smart grid including how the different layers for configuration, operation, informatics and security will connect and exchange data.
In summary, manufacturing is a vital part of a healthy economy and the competition to attract manufacturers is a global one. Other countries, including Brazil, China, France, Germany, and India are investing in R&D, and infrastructure to improve their ability to attract, support, and maintain the manufacturing industry. The U.S. should not fall behind in manufacturing leadership and the government has multiple important support roles to play. Without government help, manufacturers would end up with less efficient infrastructure and higher indirect costs. Industry, academia, and government should strive to make “Made in America” a shared pride for all Americans. More visibility on the role of government and constraints to a competitive environment can help the public understand the importance of these types of investment for the future of the U.S. manufacturing industry.
References
[1] The Government’s Role in Ensuring a Strong Transportation Infrastructure, U.S. Rep. James L. Oberstar, 2007
I am sharing below some of my favorite highlights from the round table discussions on Smart Manufacturing at the 2017 Manufacturing & Technology Conference at Cleveland. MESA International organized this forum and there were many common discussion threads and questions among manufacturers moving forward with Smart Manufacturing and IIoT initiatives in their respective companies.
After going around the table with introductions we decided to frame the discussion around a repeating story line: A manufacturer has multiple plants. They have a mix of plants running different Manufacturing Execution Systems (MES) plus some plants running on paper and spreadsheets. Some plants had over the years moved from spreadsheets to custom apps built on top of desktop database software and ended up with a few custom built MES solutions.
The manufacturer has just finished implementing ERP. The focus of the ERP project had been to standardize accounting practices and lower inventory costs. JIT initiatives drove the implementation of ERP a few years ago and they finally finished.
The manufacturer is now looking to connect manufacturing processes using a more standard solution across plants.
Why? They have a new manager and he wants to be able to compare plants and prioritize the issues he is going to tackle. Where does he start? He needs visibility of metrics across the plants and needs numbers that he can use to compare. For real transparency, standard ways are needed to collect data and turn that data into metrics that are consistent across the plants.
There is also a push to lower the maintenance cost of the various custom MES solutions. The various MES solutions are slow to be enhanced and are not keeping up with OS and platform changes.
We agreed that this was a typical scenario that everyone could relate to. We continued with more in depth questions and examples to figure out approaches to better tackle next steps by learning from each other’s experience.
Q: The ERP vendor says they can handle the shop floor also. Should we go with the ERP vendor or go with an MES vendor? Should we get guidance from vendors or from a system integrator? Consultants are expensive resources.
A: The group had seen all the above approaches. The recommended practice is to select an advisor and vendor with experience in the specific industry. Interview and survey a few and select one you are comfortable with. This approach seems to yield the best guidance and results based on project experiences presented at the conference and documented by industry analysts.
Q: What are typical first steps that companies are taking on the journey to a future smart factory? What are companies doing now to get ready for the future?
A: One company started grabbing numbers from machines and putting them into a database. Managers liked it and wanted it in more areas. They said “Just Do It” .
Q: Did you do ROI analysis?
A: Justified first project as productivity improvement. Second project justified to better fulfill traceability requirements.
Q: Is it an all or nothing strategy? Is your plan to connect all machines or do you prioritize among machines and work centers?
A: One company decided to start with new work centers and new equipment first because the equipment had better support for connectivity and APIs. However, they are hoping to connect legacy equipment in the future.
Another company started by blueprinting their as-is processes.
Q: Did you apply Value Stream Mapping (VSM) to your as-is analysis? Some of the things we are working on within MESA is to create examples of applying VSM to manufacturing automation and IT projects. We would like to create templates and examples on how to justify and prioritize these types of projects using VSM type of methods.
A: We did not use something formal like VSM, but our management likes Lean so it would probably be a good idea.
Q: Many vendors pitch: give me your data and we can analyze it. But “getting the data” is the hard part. We are being asked to be “world class”. We are getting world class machines, but we are still chasing the process with clipboards and paper. A lot of manufacturing data on paper. Any ideas on best practices?
A: First, get your network in place. The right grade of network, cooling closets, etc. Then you need actionable data—not just any data. Spend time defining the problems you want to solve and the data needed to do the respective analysis. Be ready to shift gears as you do your analysis and find that your initial hypothesis was wrong. You don’t see correlation where you expected. You might need to start collecting different data. With the right infrastructure, it is easier to shift and add new types of sensors in different places.
Q: Management has seen the price tag to do the rest of plant and they asked us to take a step back and make sure we are doing the right thing. For example, should we keep working with this system integrator, with this vendor, or should we build it ourselves?
A: We use the OT/Automation group to do these projects. We do not wait around for IT. If we could only marry the IT team’s smarts of architecture and coding practices with the OT team’s smarts on the automation needs.
Q: Have you had to deal a loss of the initial drive to connect things?
A: No. We have had a steady increase of motivation. We found that having a “poster child” deployment helped a lot. We put a focused effort into making that project very successful. Now we have developed faith with management. They believe that if we connect the machines and get data, it will lead to more savings. They have seen the results of our champion deployments. Now everyone wants to replicate that success. They want it everywhere!
Q: How can MESA help you with your initiatives?
A: We find these roundtable discussions useful. We need more forums like this. People are more willing to share in these smaller groups among peers.
I did not capture the entire discussion in these notes—just my personal favorite highlights. Let us know if you feel any of these topics deserve further elaboration in future articles. If you find these types of discussions interesting, your team should consider joining MESA International. For more information visit www.mesa.org.
Maintenance, Repair and Overhaul (MRO) operations are facing a rapidly changing highly competitive environment. In the past, MRO shops could pass the cost of inefficiencies as a “cost of doing business” to the customer, but in the face of higher competitive pressures, MRO shops must look to modernize practices and leverage digital technologies to take operations to new levels of productivity.
In this article, we discuss the convergence of technologies enabled by a new Model-Based Enterprise (MBE) philosophy that leverages the engineering 3D models throughout MRO processes including spares management, inspection and work execution. In this new digital environment, MRO shops can leverage technologies like IIoT connected product and equipment, optimized 3D illustrated work packages, inspection drones, augmented reality, and improved intelligence and analytics to create the platform for those new productivity levels.
In a recent article [2], Airbus discussed a vision for the MRO hangar of the future. In this future vision, 3D models are leveraged in many ways as foundational data for a myriad of technologies like autonomous robotic inspectors and mechanics equipped with smart glasses for hands-free computer interaction.
In this article, we further discuss six innovation areas that are driving progress towards a future MBE at leading edge MRO hangars:
3D Printing of Spare Parts
Automation with Robotics and UAVs
Asset Sensor and Diagnostic Connectivity via the IoT
Digital Thread and Digital Twin
Advanced Analytical Capabilities
Advanced Guidance with Augmented Reality and Natural Language
3D Printing of Spare Parts
The most obvious use of a 3D model is to produce the input file for a 3D printer. 3D printing is becoming a “must have” for manufacturers rather than a luxury R&D project. Especially in the aerospace and defense (A&D) industry where companies like Boeing and Airbus have been using the 3D printing process to manufacture components for more than two years, with Airbus recently printing 1,000 parts to meet delivery deadlines for the A350.
While only 7% of companies use additive manufacturing to produce their end products today, 31% use it for prototyping, and it’s predicted that 42% will rely on 3D printing for mass manufacturing in the next few years. This means that more and more CAD and CAM software systems offer capabilities to handle 3D printing.
3D printing will not only revolutionize prototyping and production, but will also transform spare parts inventory practices for service management. Manufacturers of spare parts will store the digital data to 3D print parts and components in CAD software rather than storing the physical parts on shelves. In other words, inventory will be produced only when necessary. This will be of special interest to produce hard-to-find spare parts or components that aren’t manufactured anymore.
The growth in 3D printing methods will probably come from the larger OEM and MRO companies because they can cover the upfront investment to design, build, and test the multiple prototypes it takes to finally develop a part to the desired quality level for a customer.
The companies capable of developing almost instant spare parts and completing maintenance and repair tasks within hours will have a market advantage. Imagine an airline looking for an MRO able to repair an airplane that has broken down unexpectedly. Instead of waiting days or even weeks for parts to arrive, an MRO could manufacture the parts needed outright and fix the airplane within several hours.
The ability to manufacture parts on demand will affect inventory and logistics. MROs will no longer have to keep on-hand dozens of “just-in-case” replacement components until the right client comes along. Whenever they need a part, they can just make it. This will probably cause a collapse of the entire supply chain, making some part suppliers and even manufacturers vulnerable to extinction.
Automation with Robotics and UAVs
The rapidly expanding capabilities of autonomous equipment such as UAVs and robots are being leveraged in MRO shops and some of the first applications are in automated inspection.
Examples include the use of 3D scanners mounted on robots and UAV (Unmanned Aerial Vehicle) to inspect fuselage and wings for hail, lightning, or bird strike damage. UAVs can complete detailed aircraft checks, by comparing the images captured against the respective 3D models, and quickly report any damage requiring repairs. Checks that now take 4-6 hours can be performed in less than an hour.
For now, operators control the UAVs, either guiding the inspection or setting a predetermined pattern. The UAVs fly around aircraft, providing images of skin with the same detail as visual inspections. Engineers assess damage as they are presented images. The software continues to be enhanced to rapidly focus on possible problems areas and even suggest actions based on prior recommendations. Once the UAV learns the aircraft it is to inspect leveraging its 3D model, it can recognize a known point on the airframe, and fly a predetermined inspection pattern.
LHT’s Mobile Robot for Fuselage Inspection (Morfi) is automating inspection of metal fuselages for thermographic crack detection. Morfi moves over preprogrammed points on fuselages rapidly. Two coils heat the points with electric pulses and changes are recorded on infrared cameras, revealing cracks as shallow as 1 millimeter. Vacuum pads on the robot’s feet allow it to reach vertical and overhanging sections of the fuselage. Up to four areas can be inspected and results saved in less than 30 seconds.
Asset Sensor and Diagnostic Connectivity via the IoT
It is easier to embed micro computers in many devices and assets. The device will also be expected to easily connect and exchange data with other systems and centralized data repositories that will focus on analysis of exception and aggregate data in different dimensions including dimensions relating to the asset’s 3D model. By relating the intelligence gathered to the different 3D model areas, engineers can use performance data, not only to diagnose issues, but also to enhance future designs.
Today a simple engine designed for industrial use has hundreds of sensors recording thousands of measurements, along with enough memory to store months or even years of data. Until recently, the primary use of these embedded measurements and associated low level diagnostics was to help identify component failure to facilitate the identification of areas requiring repair.
Next in the evolution to a more digitally connected future is the ability to easily connect each device or asset to centralized or remote data processing capabilities using common internet networking protocols. This type of device connectivity is referred to as the Internet of Things (IoT). Will we be able to improve product service if we can remotely monitor the product and apply advanced analytics to predict potential issues before they happen? Can we optimize maintenance based on sensor and usage statistics versus prescribed intervals?
The increased capacity and lower cost of data storage makes it possible to store huge amounts of data. But these streams of data from sensors and diagnostic feeds need to be managed. Platforms for data management includes strategies for how much to store, locally versus centrally, raw versus aggregated, and for how long for different process orchestration, record keeping and analytical purposes.
Digital Thread and Digital Twin
In addition to the asset usage data collected via IoT or other techniques, MROs need to collect as-maintained data and cross reference the respective as-designed specifications. The flow of information between engineering specifications, requirements and actual service realization records is covered under the label of the “digital thread”.
The concepts of the digital thread and digital twin have been spearheaded by the military aircraft industry and their desire to improve the performance of future programs by applying lessons learned through these digital technologies in current and upcoming programs.
The digital twin refers to a digital model of a particular asset that includes design specifications and engineering models describing its geometry, materials, components and behavior, but more importantly it also includes the as-built and operational data unique to the specific physical asset which it represents. For example, for an aircraft, the digital twin would be identified to the physical product unit identifier which is referred to as the tail number. The data in the digital twin of an aircraft includes things like tail number specific geometry extracted from aircraft 3D models, aerodynamic models, engineering changes cut in during the production cycle, material properties, inspection, operation and maintenance data, aerodynamic models, and any deviations from the original design specifications approved due to issues and work arounds on the specific product unit.
The digital thread refers to the communication framework that allows a connected data flow and integrated view of the asset’s data throughout its lifecycle across traditionally siloed functional perspectives. The digital thread concept raises the bar for delivering “the right information to the right place at the right time”.
A typical as-is condition for many MROs without a digital thread:
Design 3D models are usually not shared between OEMs and asset owners
Some drawings might be included with OEM maintenance manuals and delivered via PDF files
Maintenance requirements are documented within the maintenance manuals to be parsed manually by asset owner
Asset owner creates their own maintenance requirements based on OEM maintenance manuals and refer to specific maintenance manual sections
MRO creates maintenance task cards based on asset owner maintenance requirements and maintenance manuals
As-maintained records are archived by MRO and delivered to asset owner as PDF package. Both asset owner and MRO must maintain these records archived for auditing purposes
Progress towards a full digital thread:
Asset’s 3D models are shared by OEM with asset owner by 3D technical packages and standards like 3D PDF and STEP.
Maintenance manuals are shared using standards like S1000D for easy parsing into downstream systems.
Maintenance requirements are managed online and cross referenced to sections of maintenance manuals
Maintenance task instructions are developed leveraging links to maintenance manuals and 3D models for illustrations.
Asset’s as-maintained data is delivered in parsable data format to asset owner along with finished asset
Advanced Analytical Capabilities
Humans will manage the exceptions and will soon only need to be involved when something goes awry. Integration between operational systems, business process management systems and analytics will become the norm. There are four general levels of analytic capabilities:
Descriptive Analytics: “What has happened?”
Diagnostic Analytics: “Why did it happen?”
Predictive Analytics: “What is likely to happen next?”
Prescriptive Analytics: “How it can be encouraged or prevented?”
Descriptive Analytics
Descriptive analytics collect and report on the changing state and diagnostic properties of devices over time. This data can be used to determine stability, trends, and patterns in those properties.
Descriptive analytics also allow comparative analysis between the performance of a specific asset and the performance of similar assets in order to identify unique deviations from the asset population.
Diagnostic Analytics
Diagnostic analytics are just a step beyond purely descriptive numbers. One big problem for the IoT is the massive number of alerts that it can generate. Alerts are generally intended to get humans to pay attention, but “alert fatigue” can set in fast if there are too many of them – as there are already coming out of today’s aircraft.
Diagnostic analytics can determine whether alerts really need attention and what is causing them. It involves correlating multiple properties and looking for escalation based on the grouping of multiple conditions.
Predictive Analytics
Predictive analytics is where we start seeing bigger payoffs from diagnostic data connectivity and processing. The idea is to analyze historical data patterns that are leading to device failures, find the leading indicators, and start using those patterns to predict failure before it happens again on the same device or other similar devices.
The more data we have from similar devices or assets, the more accurate the prediction algorithms are going to be. OEMs and asset owners are creating ways to accumulate the needed data from their assets as they go into service.
Prescriptive Analytics
The initial steps for prescriptive analytics, are systems that use artificial intelligence and expert system techniques to continuously learn from human maintenance prescriptions under different circumstances, to become an automated “second opinion” for the human that proposes several options based on analyzed past decisions.
The next prescriptive level would be for the machine to start making recommendations for actions directly, with high confidence levels, from analyzing the input data. For example, an airline pilot could be told, “Shut down engine number three now, before it overheats.”
Predictive and prescriptive analysis have the potential to radically improve how maintenance activities are planned in the future. Today, most maintenance activities are planned based on a combination of elapsed time and asset usage frequency. Today, we are probably doing a lot of maintenance that is not really needed or perhaps not enough maintenance in some cases. For example, for an automobile, the dealer prescribes the frequency of brake service based on miles driven. However, the mechanic checks the brake pad usage to verify if service is really needed because the real need depends on multiple variables like driving style and the number of hills in the area. We might have taken the car in for brake service too early and would probably go ahead and get it done to avoid an additional service visit. Enhanced sensors and analytics could feed future applications that will prescribe the ideal time to take a car in for service and group multiple needed services into fewer visits. If we can see the value of this type of app for our car, imagine the value of this type of analysis to optimize and minimize service downtime for airplanes.
Advanced Guidance with Augmented Reality and Natural Language
Augmented reality (AR) or mixed reality is the ability to layer digital 3D images and virtual objects on top of the real-world images when seen through technologies such as smart glasses or hand-held tablet computers.
Multiple companies are experimenting with the use of AR for support on inspections and repairs. AR may be able to recognize the asset a user looks at, overlay the points where service is needed, and integrate with a tablet that helps the user know what work needs to be done and instructions needed. Such capabilities could improve service efficiency; lowering the cost of labor; and reducing errors.
For example, some of the automated inspection methodologies mentioned above could be further enhanced with augmented reality technology by projecting the results, color coded for severity, on top of the physical surface so inspectors and repair personnel can quickly focus to the specific problem areas.
The two major technological trends for AR display today are (1) display on mobile tablets that requires the user to hold the device, (2) display on a head-mounted display like smart glasses. Smart glasses are getting closer to the form factor of safety glasses which is needed to really make this technology appealing at the shop floor.
The traditional way to interact with smart glasses has been with simple voice commands like Next, Yes, No, Check. But in parallel to the AR technology, there is much progress on natural language capabilities for easier voice driven interaction with these devices. Chat bot technology, like seen in Siri or Alexa for home use, is being married to AR capabilities to create a new generation of hands-free easier user interaction with the advanced AR display where the user can chat with the device in a more natural manner. For example, “Glasses, show me the illustration for the next step.” The system would know what to display based on the recognized asset, the completed prior work, the work assigned to the user, and the authored work instructions.
Everyday work cards and checklists can easily be performed through these wearable devices. Steps may be signed-off as they are performed and all information systems are updated with real-time information via Wi-Fi directly from the AR device.
The above technologies and innovations are examples of how leading MROs are paving the way to the fully model-based future of MRO operations. MRO shops need to continue this innovation path leveraging 3D models as they connect and optimize processes across engineering product specifications, maintenance requirements, and maintenance execution techniques.
References
[1] Model-Based Enterprise: An Innovative Technology Enabled Contract Management Approach, Mitzi Whittenburg, Journal of Contract Management, 2012
– Using Manufacturing Operations Management Software to Reduce Cost of Quality
Tackling the price of quality goes beyond reducing the number of defects. It involves evaluating the entire quality management system. Following are some ideas to help manufacturing organizations reduce cost while improving quality levels through smarter use of integrated information systems.
Quality guru Philip Crosby wanted us to see quality management as an embedded part of everyday practice; instead of seeing quality as a separate complex process of verification, layered on top of production as overhead. Crosby described quality costs as the price of making things right the first time plus the price of making things right when they are not. Doing things right the first time, every time; is not as easy as it sounds, because many production processes have sources of variance and margins of error. To minimize the issues and defects that arise from those margins of error, we need a quality management system (QMS). Guidelines for a full QMS have been developed and documented over the last few decades by industry standards like ISO9001, AS9100 or ISO13485.
Breaking Down the Cost of Quality
The most widely accepted method for measuring and classifying quality costs is the prevention, appraisal, and failure (PAF) model which divides quality costs into the four categories in Figure 1.
The PAF model is a combination of ideas developed in the 1950’s by multiple quality gurus including Juran [1] and Feigenbaum. [2] Their concepts emphasized that it is better to have more upfront preventive investment in quality management to minimize the costs of quality failures later in the product lifecycle.
Research shows that the cost of poor quality (including rework, returns, and reduced repeat sales) can range from 15%-40% of business costs. [3] Feigenbaum referred to these costs as the “hidden factory”—the part of the factory dedicated to fixing bad work. [2][4]
See Figure 2 for examples of the breakdown of cost of quality.
The Relationship between Prevention and Failure Costs
The graph in Figure 3 shows the relationship between prevention costs and failure costs. There is a natural cost tradeoff between how much an organization spends on prevention versus how much it spends on fixing failures.
If factors affecting cost of failures and cost of prevention are static, then the only way to reduce cost it to lower the number of defects. To radically change that equation, an organization must tackle reducing the costs of the quality management system itself at the same time they are trying to reduce the number of defects.
Six-Sigma Projects Reduce the Number of Defects
A traditional way to tackle reducing the cost of quality is to reduce the number of defects. This is the focus of most Six-Sigma projects. Figure 4 shows how improving the sigma level would reduce defects and reduce the costs of the related failures up to a point. At some point the cost of quality will go up to achieve higher quality levels unless the cost of quality is tackled directly.
The Effect of Improving the QMS Processes
A different way to reduce the cost of quality is to make the processes for handling prevention and failures more efficient. Figure 5 shows how the cost curves would change when the focus is on improving the quality management system itself.
Additionally, when improving the efficiency of the quality management processes and reducing the cost of failures (cost of processing a non-conformance, cost of inspection through automation), the savings can be reinvested into better prevention methods, such as more accurate machines, better tooling, and more training, which would lead to even higher levels of quality.
The use of a Manufacturing Operations Management (MOM) or Manufacturing Execution System (MES) can make a big difference to the efficiency of quality management processes. The following are a few specific examples of how to reduce the cost of quality with improved integrated quality management processes.
Reducing the Cost of Quality Failure Prevention
Prevention costs are incurred to prevent or avoid quality problems. These costs are associated with the design, implementation, and maintenance of the quality management system. They are planned and incurred before actual assembly of the product at the shop floor. The prevention costs include specifications and inspection requirements for incoming materials, processes, and finished products, worker training and gauge calibration.
Work Instructions and Process Change Management
The manufacturing of a complex product (such as an aircraft or satellite) involves the management of a continuous stream of engineering changes directed at work in process.
On a shop floor with a paper system, changes to work instructions are written into the margins with red ink and stamped by a liaison planner. A copy of the redlined document is routed back to a process planner to get the changes incorporated into future releases of the work instructions.
An MES system can combine (1) the redlining of the work instructions on the shop floor, and (2) the update of the template work instructions maintained by MES.
Verifying Personnel Training and Certification
Personnel must be certified as competent based on education, training, skills, and experience. Personnel qualification processes must be standardized and documented.
On paper, certification of employee for the task is left for the supervisor to verify. An MES can validate each employee’s skills and certification against the latest training records before they sign on to a job.
Verifying Equipment Calibration
Equipment resources must be maintained to assure their optimal performance to capabilities, especially measurement equipment used to verify the product. The maintenance and calibration processes for equipment and tools must be standardized and documented.
An MES can verify calibration status for equipment and gauges specified to be used for data collection. Tools can be calibrated more efficiently based on actual usage instead of using the traditional date driven expiration scheme which ends up in tools being calibrated more often than is truly needed.
Reducing the Cost of Quality Appraisal
Quality appraisal activities are the most conventional quality practices and the cost of these activities are a very visible expenses because it is easy to see the cost of inspectors, testers and their equipment in the balance sheet.
Appraisal is an expensive and unreliable way of achieving quality. Appraisal in its best form is verification that the production processes and preventive measures are working. Appraisal in its least productive form, is separating the good from the bad product, counting defects, scrapping and calculating yield. While these activities are a necessary part of the quality program, it is important to move towards more preventive methods to reduce the cost of failures—the cost of poor quality.
Is 100% Inspection Needed?
Ideally the preference is to move away from 100% inspection and towards more inspection by production personnel; leaving only a small percentage of random over inspection for quality management personnel. An MES can monitor the results from inspections and the history of defects. Based on that history, process improvement efforts can focus on areas where problems are found instead of blanket improvement campaigns spread among all areas of production.
Another way to reduce inspection levels below the 100% inspection, 100% of the time, is to only trigger comprehensive inspection based on production process validation (PPV) or first article inspection (FAI) rules. In other words, when there is (a) a change in design, affecting fit, form or function of the product, or (b) a change of manufacturing sources, processes, inspection methods, location, tooling or materials, that can affect fit, form or function.
Automated Inspection and Statistical Process Control (SPC)
Where inspection is needed, and a lot of data needs to be collected, there might be possibility of reducing the clerical cost of data collection using automated inspection methods such as CMM (Coordinate Measurement Machines) or visual inspection machines.
Inspection and test results coming out of these machines can be imported directly into the MES. Critical measures and results can be tied to data collection points and to SPC run charts to monitor control levels. The X-bar and R run charts produced could also trigger alerts automatically for out of control conditions. Automation and integration may yield new levels of efficiency over the traditional processes of inspectors running spreadsheets and separate SPC software processes on the side.
Automated Quality Metrics
How many people are dedicated to putting spreadsheets and charts together for weekly meetings. This is clerical non-value added time that can be eliminated. In addition to automating SPC, systems can automate the calculation of all quality metrics off the information collected by MES software.
Process Audits
Process audits are used to confirm that the quality management system is functioning correctly. The organization can perform internal audits and external audits to suppliers as part of their periodic assessment.
Through an audit, an organization can identify a system’s ineffectiveness, take corrective action, and ultimately support continuous improvement. Process audits provide a form of assurance to management as well as regulators that your entity is following contractual and industry regulations.
Audit scheduling and check lists are an important part of the audit planning system. Audit findings should be documented and prioritized per risk and management objectives.
Reducing the Cost of Quality Failures
During inspections, it is possible to find some aspects of products, materials and components to be nonconforming to specifications. These nonconformances could lead to rework, scraping, returns and recalls all of which should be documented in nonconformance or discrepancy reports and classified in a database in such a way that the organization can use the data to determine costs and areas for improvement.
Discrepancy Documentation
The cost of filling out a nonconformance report on manufactured or purchased parts includes the cost of:
Problem Identification
Containment Actions - activities to ensure nonconforming material is controlled and prevented from improper usage
Correction and Disposition Documentation – instructions to correct and bring the product back to conformance, and instructions to handle component scrap or repair if needed
The use standard codes and descriptions for symptom, defect and cause types, can simplify and speed up the documentation process and provide more consistent data for analysis.
Workflow Driven Routing of Tasks
The costs of routing the discrepancy for disposition through a workflow process across multiple departments can be controlled by limiting the participants in each discrepancy to those who are needed instead of including the entire multi-discipline Material Review Board (MRB) in every review.
Rework, Re-inspection, and Re-testing
One of the advantages of using an MES is the handling of rework instructions to correct an issue. The same process planning tools used for authoring regular work instructions are used for authoring rework instructions and either append work to the original work order or edit the instructions for the affected units only. The resulting rework jobs should show on the same job list along with the planned work so the mechanics do not have to learn a different process for rework, re-inspection and re-testing. However, these activities do need to get classified, and segregated in reporting to the financial system as cost of poor quality.
Corrective and Preventive Actions (CAPA)
When failures do happen, the team should assess if adjustments are needed to the preventive measures to avoid recurrence for the same type of problem. This is done through the Corrective and Preventive Action (CAPA) process.
The cost of processing a corrective action includes the following activities:
Problem Identification including recurrence data, cost impact and risk classification
Analysis for root cause by a team of subject matter experts
Implementation of an action plan to correct the root causes identified
Verification tasks to make sure that the plan was implemented and effective
Do not flag every little problem for a full CAPA process. Select only problems that seems to be recurring and have a significant impact. This can be figured out by ranking issues based on a standardized risk assessment methodology.
It is also important to check history for similar problems and see if there have been solutions that have worked in other areas.
The Cost of External Quality Failures
The intangible costs of external quality failures, including customer dissatisfaction, loss of reputation and loss of future sales, might be hard to calculate but are not hard to picture as having a huge negative impact on the future of the business. The best way to reduce external failure cost is to not have it at all! The best way to avoid external quality failure costs is to focus on improving the other three costs of quality.
Conclusion
There is a natural cost tradeoff between how much an organization spends on prevention versus how much it spends on fixing failures. In addition to the traditional way of reducing the cost of quality by reducing the number of defects, it is possible to tackle the efficiency of the quality management system itself to further reduce cost of quality. The use of MES software leads to more embedded quality control throughout the entire manufacturing process, more ownership of quality by production personnel, and more smiles and fewer tears for everyone in the organization.
References
[1] “Quality Control Handbook – 1st Edition”, Book by J. M. Juran, McGraw Hill, 1951
[2] “Total Quality Control”, Book by A.V. Feigenbaum, Harvard Business Review, 1956
[3] “A Review of Research on Cost of Quality Models and Best Practices”, Whitepaper by A. Schiffauerova and V. Thomson, International Journal of Quality and Reliability Management, Vol.23 No.4, 2006
[4] “Dr. Armand Feigenbaum on the Cost of Quality and the Hidden Factory”, Article by Tim Stevens, Industry Week, 1994
[5] “The Six Sigma Revolution”, Book by George Eckes, John Wiley & Sons, 2001
It is easy to come out of a presentation on “big data” these days thinking that the future of manufacturing operations management and manufacturing IT is a combination of three things: connected machines, data stored in a “data lake”, and an analytical engine. However, those closer to manufacturing systems management know that the reality of data storage and processing needs are vastly oversimplified by drawing a big data lake circle in the center of a diagram. In a prior article, “Unraveling The Hype On Data Lakes for Industrial IoT” [1], we discussed some of the myths around data lakes. In this article, we dig deeper into data processing and examine new and older technologies under the lens of the requirements to: distribute data from its source to departmental functions, consolidate views of data for multiple organizational levels, along with new processing opportunities presented by data streaming technology.
The traditional data landscape across an enterprise is a myriad of islands of information with varied data structures and indexing schemes controlled by each department’s applications for functions including Procurement, Inventory, Production, Quality, Human Resources, Sales, Marketing, and Finances. Rather than trying to create a single mega system to manage the enterprise–one designed from the ground up with a consistent and integrated data model–organizations need ways to integrate, exchange and manage data across the departmental functions. There are different methods and technologies available to perform data distribution, processing and consolidation tasks including:
Enterprise Application Integration (EAI)
Extract, Transform and Load (ETL)
Enterprise Information Integration (EII)
Data Stream Processing and Streaming Analytics
These techniques should be viewed as complementary with different functions in an overall IT technology stack. In fact, many vendors are combining multiple of these technologies into unified platforms in recognition of how they complement each other in different applications.
Enterprise Application Integration (EAI) - Application-to-application integration via messaged-based, transaction oriented, brokering and data transformation. Focuses on integrating business-level processes and reacting to events that occur during those business processes with transaction data messages for handoff to other business processes or systems. EAI integration mapping needs to be updated when applications change their integration interfaces. It can have simpler results than other methods for the end user because data is accessed via their main departmental application and integration is handled in the background by the IT department. Hub and spoke methods using middleware like Enterprise Service Bus (ESB) products facilitate message brokering between multiple applications.
Extract, Transform and Load (ETL) – Set-oriented point-in-time transformation of data for migration, consolidation and data warehousing. It can process very large amounts of data during non-peak use times and deliver data optimized for reporting, while freeing up transactional databases from the computational burden of complex queries. ETL scripts need to be updated when applications change their query web services or transactional data structures.
Enterprise Information Integration (EII) – Transparent data access and transformation layer providing a virtual consolidated view of data across multiple enterprise data sources. In other words, EII provides federated data queries across multiple data sources via one meta model. EII models can be placed on top of data warehouse models and join structured and unstructured data. It can provide real-time read and write access and manage data placement for performance, accuracy and availability. EII can be hard to implement when key fields and data types do not match across source data systems. EII can impact network bandwidth and have high demand on source systems during peak hours. Caching schemes improve performance by extracting data out of data source systems for reuse by multiple EII queries. The Hadoop toolkit is an example of a technology stack that enables EII techniques.
EII techniques are usually leveraged by data scientists for ad-hoc analysis while ETL techniques turn data into simpler structures for every day reporting needs of users that don’t need to understand the complicated data structures of source systems.
Data Stream Processing and Streaming Analytics
In contrast to the traditional database model where data is stored, indexed and subsequently queried, stream processing takes the inbound data while it is in-flight real-time through an EAI process and is able to look for alert conditions and patterns in the data stream or in-memory data aggregation—the “real-time data mart”.
There are some evolving combinations of the above techniques and technology stacks which use data stream processing techniques as a pre-processor during the extract phase of an ETL process before data ends up transformed and stored at the data warehouse.
Data in the Smart Manufacturing and IIoT IT Framework
Evolving Smart Manufacturing and IIoT systems and applications leverage these different data handling techniques to improve real-time analysis where needed, as well as provide improved availability of structured and unstructured data for different types of reporting and analysis where real-time data is not needed or even desired. For example, the weekly review by the management team of Key Performance Indicators (KPIs) for the organization should be based on data mined at a fixed time—not data mined real-time at different random times yielding different results for every query. We don’t want each manager showing up to the weekly meeting with a different set of data in their hands. In contrast, we want to know of a production machine about to go down as soon as possible especially if it will stop the production line.
Figure 2 shows an example of how we could put all these techniques to work together in an IT architecture for the enterprise connecting different types of data for different types of data processing applications. The reality for this picture at any one organization might have more additional parallel paths but this simplified view illustrates the general idea of how it can all work together.
The picture gets even more interesting in Figure 3 when we add the fact that we want to connect the entire value chain for a product including enterprise data from multiple manufacturing sites and operational data from multiple tiers of suppliers across the globe.
As we look forward to cloud integration we expect the data services to support integration to multi-tenant cloud applications like Salesforce and support for REST and JSON APIs.
The hype about data lakes, big data and all the great data generated by the IIoT (Industrial Internet of Things) keeps growing. There are valid use cases for data lakes in manufacturing operations management and the Smart Manufacturing IT framework but companies should know more about the realities in this arena. Gartner recently published a report titled “Three Architecture Styles for a Useful Data Lake” [1] which mentions these concerns. We need to make sure they don’t stay buried in that report and instead bring them to the front of the conversation.
First, what are data lakes?
A data lake is a collection of storage instances of various data assets stored redundantly in a near-exact copy of the source format. The general idea is to store data separate from source domains for future processing and analysis by different applications in the enterprise.
A data lake is a concept, not a technology. There are multiple types and purposes for data lakes. There could be more than one data lake in an IT architecture. Data lakes may add contextual metadata to the raw data. They may have a short or long shelf life for the data depending on the purpose for each data lake.
Examples of the use of data lake concepts range from self-service intelligence to ad-hoc data science analysis to advanced analytical alerts on new data trends. However, the data lake is not the new data warehouse. It is a place to search for and discover additional insights, it is not optimized for daily or weekly reporting of metrics to the organization.
Second, what are the misconceptions about data lakes?
There are many valid reasons to consider data lakes in the information technology landscape for future manufacturing systems. However, they should not be viewed as a silver bullet answer to the complexity of architecting and integrating enterprise platforms. They should be viewed as another tool in the IT arsenal, to be used where appropriate, and approached with an understanding of the following myths.
Myth: A data lake strategy does not include ETL tools
Key to the data lake concept is the separation of independent data storage and computing software. The methods of moving data into and out of the data lake can include tools like ETL (extract, transform, load) and ESB (enterprise service bus).
Myth: A data lake has to be Hadoop
Hadoop is the most popular platform for data lakes. However, it is possible to use other types of relational, NoSQL and object data storage tools. Some of the factors to consider in this decision include data volume, velocity, variety, and scalability.
Myth: Data lakes are inexpensive to implement
Even though data storage cost continues to decrease, the costs of networks, integration, backup, security and support can add up very quickly and significantly. Open source software and cloud storage options can help with cost but they should not be viewed as free of cost.
Myth: Get all the data you can into the data lake. No need to worry about data modeling.
Gartner predicts that by 2018, 90% of data lakes will be rendered useless as they become overwhelmed by a vastness of data captured for uncertain use. The concept of garbage-in-garbage-out applies to the data lake. Data hoarded without a preplanned purpose or proper contextual or lineage information will create a lot of noise that can obstruct finding important signals in the data.
Myth: A data lake creates a single source of truth
Data collected in a data lake is usually schema-on-read which means that the structure is determined through discovery at the time it is analyzed. The data lake must remain flexible to be inclusive of new data sources. Data lakes have soft governance approaches to avoid anarchy in the data lake, but they don’t have hard governance. The data lake should not be considered a system of record.
Myth: Anyone can use a data lake
It is not trivial to do data discovery and model development in the data lake. Organizations will need staff skilled in statistics, machine learning and data science to reap benefits from data lake efforts.
Myth: A data lake is a data integration method
The data pipelines in a data lake scheme should be viewed as additional-to and not in-lieu-of data integration schemes linking the organization’s business systems. It is possible to leverage data integration messages as one of the sources for a data lake, but the data lake purposes should always be secondary, and should not interfere with the queuing and synchronization required for system integration purposes.
Myth: If we build a data lake, people will use it
The implementation of technology should always have specific use cases and a business purpose in mind. If we start with a need, the system will be used. If we start with technology and the need is not clear, there is a big risk that no one will fish in that data lake.
References
[1] “Three Architecture Styles for a Useful Data Lake”, S. Sicular, Gartner, 2016