How manufacturers can capture and codify the knowledge of experts

Manufacturers need a better way to collect, organize and provide access to knowledge that’s vital to boosting their workforce’s capability.

In brief

• Challenges in accessing critical job-related information prevent a manufacturer from being able to consistently build the workforce’s knowledge.
• Relevant information should be stored in an integrated multimedia database housed on a cloud platform, such as Microsoft Azure, accessible across the enterprise.
• By embracing digital solutions to enable pervasive and fast access to the information, manufacturers can raise the level of all workers’ knowledge and skills.

As manufacturers redouble their efforts to infuse greater resilience in their operations, they’re recognizing why it’s so critical to capture and codify the knowledge that’s in the heads of their experts — long-time workers who have built up significant expertise in “how things work” over time. Institutionalizing such knowledge is vital to keeping the shop floor running, especially in times of disruption.

But that knowledge isn’t very useful if other workers can’t benefit from it. In this article, we look at how digital can help manufacturers make experts’ knowledge accessible (see figure 1) to the broader workforce to improve their skills and performance, as well as boost the enterprise’s overall resilience.

The challenges of access to knowledge

Today, it remains commonplace for manufacturing operations to be overly dependent on a small group of experts for deep technical knowledge. In one manufacturing company we worked with, technical equipment expertise was limited to just three people globally. These centrally located, long-time employees spend most of their time flying from site to site helping less-experienced workers learn how to operate and fix the machines used in production, but never really upskilled those workers sufficiently. 

This isn’t exactly an optimal model even during normal times. Its inherent inefficiencies and non-scalability make it difficult for plants to keep their lines operating at peak performance, and also makes it tough to quickly bring machines back up to speed when they go down. In times of severe disruption, the model breaks down entirely. For example, with travel restrictions related to COVID-19, such experts suddenly can’t travel to where they’re needed, which makes production vulnerable to risk.

Of course, manufacturers do have other repositories of knowledge beyond these tenured experts. Most organizations have a shop floor production system, which includes daily routines, standards and procedures. These may be printed manuals and documents for relevant machines or databases containing vast tomes of information. There also may be local SharePoint sites where plant-specific information is stored. But navigating through this information locally is a manual and time-consuming process that lacks a clear structure and path to what a worker actually needs to do his job. And sharing such information across the enterprise is pretty much out of the question.

Challenges in accessing critical job-related information don’t just make it difficult for workers to get what they need to deal with a specific situation (such as how to do a changeover or set specific centerlines), they also prevent a manufacturer from being able to consistently build the workforce’s knowledge and skills on an ongoing basis. This is also a growing obstacle to recruiting digital-savvy younger workers who don’t find working in a heavily paper-based environment all that attractive.

The fact is, manufacturers need a better way to collect, organize and provide access to the information and knowledge that’s vital to boosting their workforce’s capability and, by extension, enhancing the enterprise’s overall performance and resilience. Digital tools are paving way for this development in a growing number of leading manufacturers.

An integrated database: making situation-specific knowledge more accessible and widespread

Day-to-day shop floor workers find that they need access to different types of knowledge depending on the work they’re doing. Most of what workers do typically fall into two buckets: routine and regular tasks, which are repetitive and occur with a predictable frequency as a result of the organization’s standard work practices; and routine and non-regular tasks, which tend to be more complex, such as product changeovers or quality inspections.

Generally, the vast majority of relevant information for these tasks can and should be stored in an integrated multimedia database housed on a cloud platform, such as Microsoft Azure, that is accessible to anyone across the enterprise. Able to handle both structured and unstructured data, this database contains digitized versions of all the printed manuals, standards, procedures and other documents manufacturers currently use, as well as short videos that clearly show how to complete tasks workers would commonly be asked to do. A powerful back-end search engine enables workers to quickly and easily look across the entire database to find the information they need. It’s a far superior alternative to having to read through pages and pages of manuals or scour hundreds of files on a SharePoint site, and much more in line with how people today want to consume knowledge and information.

But that’s just the first step. Once a manufacturer has been able to digitize and centralize all its knowledge in one place that’s much more efficient to search through, it can move from self-service to assisted access. This means adding an artificial intelligence (AI) layer such as Azure Machine Learning that actually serves up what a user is likely looking for, based on many similar past interactions — and it does so in a fraction of the time the user would take to find the information on his own.

This technology recognizes or learns, for instance, the types of problems workers have encountered, how they were typically addressed, and which interventions were the most effective and which weren’t. By guiding a worker through this type of analysis and problem solving — often through a chatbot or a voice-enabled digital assistant — a machine learning (ML) engine can help accelerate problem solving, giving the worker confidence that he’s taking the right action to deliver the desired results. When ML is paired with translation technology such as Microsoft Translator, a manufacturer can streamline access to knowledge for all workers, regardless of the location and the languages they speak.

While this approach to knowledge access has yet to permeate the shop floor, it’s well advanced in other industries and applications. It’s played a big role in helping companies transform their customer service organizations by giving both customers and agents easy access to knowledge and information. The same technology and approach used for AI-powered customer service, which helps clients and agents more quickly and accurately solve customers’ problems, can be used to help shop floor workers get answers to the questions and challenges they encounter every day.

ML’s use is also growing in the medical field. Some leading health care providers are pooling their doctors’ knowledge into a central database that ML uses to diagnose illnesses based on patients’ symptoms. The more the system learns, the more accurate it becomes — to the point at which it can rival or exceed human doctors. If this approach can be used to diagnose patients’ complex illnesses, it certainly can help identify the root cause of a problem on a production line that’s behaving in a certain way.

Virtual reality and augmented reality: engaging with experts remotely

But as powerful as these solutions can be, sometimes workers still need to engage directly with an expert to solve a problem. This is generally the case with a third type of work that shop floor employees may encounter: non-routine and non-regular tasks, which are responses to events rare enough to not have associated standards. Here, technology can also help by overcoming the inherent limitations of expert scalability and availability. Collaboration tools such as Microsoft Teams and even simple video chats via mobile phones can enable experts to provide remote assistance to workers on the shop floor who encounter a situation they can’t resolve on their own. With travel being so restricted, many manufacturers are accelerating their use of these tools. But something much more powerful is emerging.

As virtual reality (VR) and augmented reality (AR) technologies continue to advance, manufacturers have an entirely new way for workers and experts to interact. For example, with Microsoft Dynamics 365 Remote Assist and the Microsoft HoloLens mixed-reality headset, workers can solve problems in real time by sharing what they’re seeing on the floor with experts in remote locations, who can then diagnose issues and walk workers through the steps to fix them.

In addition to getting workers the information and guidance they need when they need it, this solution saves money and “wear and tear” on experts, whose knowledge can be digitally “replicated” across different plants or geographies without them having to leave their home base. This is even more critical when flying to a site isn’t possible. Even without involving an expert, this solution can deliver key information from the knowledge database to workers on what they should do — thus enabling less-experienced workers to effectively complete the task at hand.

While such applications are still in their infancy, they’re not science fiction. Leading manufacturers are beginning to deploy them on the shop floor today, with maintenance being a prime use case. More is certainly to come as the technologies continue to mature.

The digital coach: providing ongoing training and development

In addition to helping address situation-specific knowledge needs, digital solutions have tremendous potential to revolutionize ongoing worker training and development. One such solution is the Smart Deployment Console which is part of the EY Smart Factory solution. The console acts as a “digital coach” by providing access to content workers, who need to continue to learn new skills, without having to depend on an individual mentor or expert. 

Based on the Microsoft Azure cloud platform, the console follows the concept of PDCA — plan, do, check and act. The starting point is a self-service assessment to understand how mature the manufacturing teams are, relative to the skills they need to do their jobs. Based on the answers teams provide to a set of maturity-based criteria, the console identifies gaps in maturity and, in turn, prescribes a sequence of activities to close those gaps. This functionality mimics the role of a coach in a digital world.

Along with these recommendations, the console delivers the specific content associated with each improvement activity, and continually maps the progress teams are making along the capability maturity curve to ensure they’re on track toward the desired outcomes (and identifies corrective action if they’re not).  Furthermore, it can correlate learning progression with manufacturing performance to track whether the learning is delivering the expected results.

Such digital coaching solutions can help manufacturers dramatically increase the consistency and level of their shop floor workers’ knowledge and skills. And it provides an environment that’s more conducive to younger workers who prefer to access information and learn. This helps manufacturers attract and retain scarce talent.

“Today, workforce transformation has become even more important for manufacturers,” notes Neal Meldrum, Business Strategy Manager at Microsoft. “The challenges to reactivate factories in a COVID-19 era require best-in-class digital solutions to upskill workers and make them more agile and safe. Together, Microsoft and EY teams have worked to leverage our technology platform to bring innovative solutions to market that address such critical business needs.”

Making progress toward “everyone’s an expert”

As the saying goes, a chain is only as strong as its weakest link. In the manufacturing world, those “weakest links” are the less-experienced shop floor workers who don’t intuitively know how to respond when issues arise. Their lack of experience and insights can prevent the larger enterprise’s operations from performing at a consistently high level and undermine the enterprise’s ability to bounce back from disruptions.

By embracing digital solutions to enable pervasive and fast access to the critical information that workers need to do their jobs, manufacturers can raise the level of all workers’ knowledge and skills — and, in the process, minimize production risks and maximize their facilities’ overall performance.

Keep an eye out for the next instalment of this series which will look at speeding decision making.

The views of third parties set out in this publication are not necessarily the views of the global EY organization or its member firms. Moreover, they should be seen in the context of the time they were made.