Define maintenance data value to gain IIoT benefits

Maintenance offers great potential value in the Industrial Internet of Things (IIoT). Getting the right data to the right worker at the right time can head off potential problems and point the way to more uptime.

By Plant Engineering, Chris LeBeau April 4, 2016

Maintenance offers great potential value in the Industrial Internet of Things (IIoT). Getting the right data to the right worker at the right time can head off potential problems and point the way to more uptime. In the first of a two-part discussion, Chris LeBeau, global IT director at Advanced Technology Services (ATS), talks with Plant Engineering about how manufacturers can leverage IIoT to greater benefit.

Plant Engineering: What do you see as the potential for IIoT as it relates to maintenance?

LeBeau: IIoT technology represents a significant addition of capabilities to existing plants and an important component of new plants. The technology, however, is only the right answer relative to the right questions. The questions are unique to each implementation. Today, there is no one-size-fits-all scenario. The unique circumstance of each plant, line or machine has to be considered.

There are some questions that do generally apply to most plants. What machines fail most frequently or have the most impact on the operation? What conditions or indications would be helpful to monitor on those machines to prevent unplanned downtime? What preventive maintenance procedures require regular meter readings or measurements performed by maintenance personnel? What information indicates the health of the operation or can provide a leading indicator of any performance degradation?

The potential benefits of IIoT for maintenance come in several forms. The overall situational awareness of plant operations and equipment condition can be greatly improved. Information can be presented in summary view as dashboards or in detailed views with real-time measurements and performance trends over time. Automated alerts based on degraded performance can provide time to act, preventing a more serious failure, or to schedule maintenance that avoids disruptive downtime.

The labor required to collect meter readings and measurements can be reduced or eliminated with sensors providing a constant flow of data versus periodic readings. Technicians could then spend that time on activities with more value to the operation. IIoT data can also be accessible outside of the plant. Centralized resources would have the ability to participate in maintenance, providing support or advice remotely. Data from like machines across several locations can be aggregated. The combined data can create a clearer picture of machine performance and be used to improve maintenance plans.

Benefits from IIoT can only be realized if the tangible value of the data to a maintenance organization and overall manufacturing operation are clearly defined. Failure modes need to be identified along with realistic costs of downtime. The necessary measurements and sensor types need to be specified along with other observable information useful in improving situational awareness. The cost of repetitive labor tasks should also be factored. Then, any estimated benefits can be tested against the costs in a business case with realistic and measurable objectives for each scenario.

PE: That said, what are the barriers to adoption?

LeBeau: Any significant growth of IIoT and sensor connectivity within plants will be dependent on the establishment of recognized standards. There is currently an overall lack of industry-supported IIoT standards. This creates risk in choosing solutions that will have broad support and long-term usability. There is also a low number of sensor options supporting the wide variety of required measurements robust enough to operate in harsh industrial environments.

Many plants are not sufficiently networked to support the addition of hundreds of new sensor end points. Wide IIoT adoption will require cost-effective connectivity. Scaling traditional Ethernet and Wi-Fi technologies to support this requirement can rapidly become too costly. Options based on less expensive radio-networking technologies that leverage unlicensed spectrum (900 MHz and 2.4 GHz) offer an alternative. Although also requiring agreement on standards, these technologies provide higher device density and longer range than Wi-Fi supporting indoor and outdoor applications.

In all cases, IIoT is a new investment in existing plants or an additional investment for new plants. Careful consideration has to be given to the overall maintenance plan and the specific-use cases where sensor data can provide returns. This is more likely in process plants where there are equipment dependencies and a higher cost per downtime hour, and obviously less likely in discrete manufacturing environments.

PE: Specifically, we hear about change management being a key challenge for plans to realize the full benefits of IIoT. How will maintenance operations have to fundamentally change?

LeBeau: This change is often characterized as the need for coordination between operational technology (OT) in the manufacturing plant and information technology (IT). Traditionally separate, these two disciplines will need to cooperate to realize the benefits of IIoT.

In the IIoT model, sensors will generate data that is collected, contextualized, analyzed and ultimately sent as actionable information into applications supporting numerous operational functions—including maintenance. Changes in monitored conditions will generate alerts and status notifications, and drive automated actions to maintenance personnel.

To keep this information relevant, IT will need real-time awareness of activity in the plant. This includes production schedules and notification of all planned and unplanned maintenance actions that could affect monitored data. Without this awareness, alerts and automated responses will cause false alarms and wasted effort by personnel unnecessarily reacting to events.

Maintenance personnel and the applications they use will have to provide context and create this awareness. Maintaining an accurate production schedule will account for times when the machines are not running or require different conditions to be measured. The start and stop times for maintenance actions will have to be indicated. These time frames will also have to be factored into calculations related to operational metrics and historical trend analysis.

All of this will require changes in operational behaviors for the maintenance personnel and increased coordination with their IT counterparts. In return, IT will have to become much more savvy about how OT functions and continually work to provide easy-to-use applications that supply maintenance personnel with relevant and actionable information. The aging workforce dilemma in maintenance has been well documented. Modernization in the manufacturing sector and technological changes like IIoT combined with a younger maintenance workforce will start to close the gap between the OT and IT domains. Over time, the distance will shrink through shared experience and knowledge transfer, ultimately becoming a significant and complementary overlap.

PE: What’s your sense of maintenance data being captured today, and how it is used? Are there still some manufacturers stuck with clipboards and paper?

LeBeau: Most maintenance-related data is collected, stored and only used locally. Accuracy is also dependent on the person or persons entering the data. In most cases, the information is maintained as separate forms or records within a computerized maintenance management system (CMMS) and not combined in any meaningful way for trending or analysis.

There is still a significant amount of this data being maintained on paper. In many cases, this works well for the people at a plant, but it also creates risk. Data quality can be impacted by personnel changes and paper records can be damaged or lost. The data is also not readily usable by anyone not actually holding the document.

PE: Look five years down the road. How will an effective maintenance department operate?

LeBeau: The reality of plant maintenance is that it is, and will remain, a physical and mechanical occupation. Computerization and automation will continue to expand inside plants of all kinds, but maintenance will still look a lot like it does today. The conversation about maintenance, however, will be much different. How it is planned, when it is done and how effective it is will all be based on the vast amounts of data coming from technologies like IIoT and the applications that collect, correlate, analyze and leverage that information.

Centralized resources will play an integral role, providing access to data and leveraging remote tools for coordination with onsite resources. This shift will provide options to have advanced skill sets and capabilities available at any plant, not just the ones that are large enough to afford people onsite. This network effect will extend to a vast array of new resources including access to knowledge and suppliers providing instant parts creation onsite when combined with 3-D printing.

Effective maintenance departments will embrace these changes to focus their work effort on the things that matter most. This will be driven by plant management having increased transparency, allowing better decisions to be made on what to do and when to maximize productivity and profit. These capabilities will also demonstrate the significant business value of maintenance and create better alignment within the operation.