ML Journal

Aligning the solution framework, intended business value, and smart factory expertise is key to success. 

Factories striving to get farther down the Manufacturing 4.0 path know they have so much to gain. The question is how efficiently and effectively they can get there. Smart, digital factory solutions such as machine condition and process monitoring sensors, machine learning (ML), and intelligent analytics are available to streamline and optimize legacy manufacturing processes — from operations and maintenance to the supply chain and facilities management. Smart maintenance initiatives are a prime example.

By identifying and capturing unleveraged Big Data and honing it into meaningful, actionable insights, plants gain greater situational awareness and the ability to make more intelligent, proactive decisions. The newfound real-time, future-focused insights accelerate actions that improve performance, productivity, cost efficiencies, and reliability, in addition to increasing safety, cybersecurity, and regulatory compliance. Protecting legacy equipment investments with this approach provides significant cost advantages over sweeping factory modernization.

“Protecting legacy equipment investments with this approach provides significant cost advantages over sweeping factory modernization.”

 

There is much to consider when taking this approach. Technology and resource challenges must be overcome, risks and opportunity costs must be weighed, priorities must be set, and a framework for success is needed. Achieving all this while still meeting current operational goals is a lot to take on at once. It is why factories are increasingly turning to trusted expert partners for assistance.

Barriers and opportunity costs

Effective data acquisition, storage, analytics, and application are central to any smart factory initiative, whether upgrading legacy equipment with integrated condition monitoring solutions, or more generally trying to make better, more timely decisions. But common factory challenges impeding the journey must be overcome, such as:

• They are insufficiently networked to support new technology at scale
• They have legacy equipment that could be a cybersecurity risk if connected to the Internet
• They do not realize the prevalence of valuable yet hidden data
• They do not really have a way to manage and use the data effectively

Beyond issues like these, there are opportunity costs to weigh for any upgrade. For example, replacing or upgrading a machine or line within a plant requires stopping manufacturing in that area for some amount of time. Hence, strategically augmenting and improving what is already in place is often more valuable than doing an upgrade, considering the longer an asset or line keeps running, the higher the return on its investment.

Additionally, tackling any transformation or smart initiative requires a well-planned, structured activity sequence. For instance, it is necessary to address the core challenges around connectivity and security before going after the needed data.

“The challenge is knowing what data is needed and why, and how it will help to do something not currently possible.”

 

The pursuit of hidden data within the factory also needs structure. Much like finding important messages in an overloaded email inbox, having excessive sensors and alerts makes it difficult to identify which issues are most important to address. The challenge is knowing what data is needed and why, and how it will help to do something not currently possible. Providing context and aligning that with other data will surface the issues that most require attention, so that the right equipment is monitored, and appropriate actions are taken.

Framework for success

The optimal approach to minimize project risk is with an Industrial Internet of Things (IIoT) solution framework focused on evolution rather than revolution. Most plants cannot afford the costs or disruption of jumping from very old technology to something very new. Likewise, just because an asset can be monitored and generate alerts does not mean it should.

Take, for example, a goal to improve asset reliability with intelligence to reduce equipment failures and costly unplanned downtime. In its report Accelerating Industry 4.0 Through Remote Monitoring and Diagnostics, IDC, a global marketing intelligence firm, observes that the average cost per hour of unscheduled downtime in manufacturing is over $110,000. Conventional equipment maintenance is time- or usage-based, regardless of actual operating conditions, and some equipment is allowed to run to failure. Smart maintenance is proactive and predictive based on condition monitoring data, which improves asset reliability and uptime with measurable ROI and increases productivity and plant performance.

The smart maintenance journey leverages digital sensor technologies that capture legacy equipment condition data for remote machine health monitoring. Even the smallest variations in vibration, temperature, oil quality, or motor function can indicate a failure is emerging or imminent. Secure, rugged, integrated sensors allow for integrated asset monitoring, data collection, processing, and storage — bringing previously hidden machine data to light from which useful intelligence can be gleaned.

Processing this data with predictive analytics and ML allows for real-time alerts and prescriptive actions to avoid asset failure and unplanned downtime. Meanwhile, failure trends can be leveraged to shape purchasing decisions, depreciation rates, and replacement forecasts.

“The optimal approach to minimize project risk is with an Industrial Internet of Things (IIoT) solution framework focused on evolution rather than revolution.”

 

The framework defines the required functional components of the solution, ranging from how the data is acquired, transported, processed, consumed, and secured to deliver business value. Framing a solution that is cost effective, capable of achieving goals, and increases business value involves:

1. Defining the journey, including what problem must be solved
2. Making the business case for pursuing it
3. Determining what steps will have the most impact
4. Building the foundation for the next steps

This process acknowledges the real difficulty of having a large and growing choice of smart solutions on the market. Selecting a functional framework that aligns with today’s problems but can also accommodate future problems will help an organization avoid dealing with multiple competing solutions. In essence, doing something similar but separately is less valuable than having an integrated smart solution that can be expanded and extended over time.

Ultimately, an effective framework will deliver a solution that is feasible, creates business value, and will support long-term strategy with maximum impact. It could even provide a catalyst for new business models, new customer engagement models, and new or expanded markets.

Value-added partnerships

Due to chronic industrial skills gaps and time and resource constraints, many manufacturers are engaging partners to help them meet smart factory goals, rather than going it alone. A recent Forrester research study, Data-Driven Decision Making Drives the Need for IIoT/Remote Sensors, completed in partnership with Advanced Technology Services (ATS), found that 55% of the respondents expected to use a combination of outsourcing to an external partner and in-house resources for data collection and analysis within the last few years. Another 27% planned to fully outsource the data collection, analysis, and gathering of insights to a partner.

“The most capable partners are those with deep technical and analytical expertise, bolstered by experience working across industries”

 

Choosing a Manufacturing 4.0 partner and their responsibilities depends on what skills and experience will best meet the needs. In the case of smart maintenance, any or all of the following roles may be desired:

• Defining condition monitoring priorities and strategy
• Implementing the right technology and communications
• Storing and managing the data
• Understanding how best to analyze and use the data

The most capable partners are those with deep technical and analytical expertise, bolstered by experience working across industries with many different types of manufacturers and equipment. They may offer a pool of experts operating from a central hub for data-driven remote monitoring and diagnostic support, in addition to regional teams of multi-craft field technicians for on-site maintenance expertise. This distinctive value proposition accelerates the time to value for smart factory initiatives.

Consider sharing the load

Manufacturers have every reason to pursue the high-value advantages of Manufacturing 4.0, such as lowering the costs of operation, increasing asset availability, and reducing unplanned asset downtime. But with the manufacturing environment and available technology solutions continuing to rapidly evolve, the complexities of implementing and expanding smart factory initiatives are growing. Factories struggling to prioritize and achieve their goals would do well to consider engaging a highly qualified partner for assistance.   M

About the author:

Chris LeBeau is the Global Director of IT at Advanced Technology Services, Inc., working with industrial maintenance experts and technology leaders to enable Manufacturing 4.0 strategies and maximize the value of technology to manufacturers. Chris previously held positions at Cisco Systems, AT&T, IBM and began his career in satellite communications with the U.S. Army Space Command.