IIoT Spotlight Podcast EP 047: Understanding the State of, and Challenges in IoT Implemention in Heavy Asset Industries – Dave McCarthy, VP of IoT Solutions, Bsquare

*This episode of the Industrial IoT Spotlight podcast is sponsored by PTC

In this episode of the IIoT Spotlight Podcast, we discuss predictive maintenance, how to use best practice behaviors to drive best practice results, and predictive maintenance technologies. We also discuss framework of balancing the three-legged stool (people, processes, and technology), and 2 case studies of successful predictive maintenance implementation projects using the ThingWorx Industrial IoT platform.

Takeaways:

1. Predictive maintenance is a method for machine operators to get the maximum productivity out of any piece of equipment. It is especially important for asset intensive industries.
2. Predictive maintenance is the basis of digital transformation in other areas of the enterprise - it is not valuable to optimize a process with chaotic variables.
3. The main value driver in predictive maintenance is the cost of production per unit produced because it is an indicator of machine reliability, revenue, and operating costs, which all add up to the profitability of the production plant. Other KPIs to be considered when implementing a predictive maintenance plan are safety, environmental impact, and product quality.
4. To determine if you are ready to adopt predictive maintenance, people and processes should be in place before the technology is implemented, to create a roadmap to build a balanced three-legged stool.
   1. People: the right people and skills in the right positions
   2. Processes: the mechanisms to identify maintenance tasks and the process to manage these tasks
   3. Technology: ability to gather equipment health data
5. Technology driving the adoption of predictive maintenance are the ease of connectivity and data analytics software tools:
   1. Connectivity increases the ability of systems to pull data out of control systems while maintaining their security. The ability to add more sensors without adding significantly more cabling of infrastructure allows more parameters to be measured for a more holistic view.
   2. Software for data analytics organizes the data into events and patterns using natural language processing, which makes the system more user friendly for the connected worker.
6. The decision to choose the software and hardware products depend on the understanding of business and operational processes:
   1. Choosing hardware: The failure mechanisms for each piece of equipment should be considered. The sensors should be able to measure the failure mechanisms and collect the data that can help the organization to plan scheduled maintenance when it is operationally convenient.
   2. Choosing software: The software must bring together business processes and maintenance processes to orchestrate maintenance activities effectively. At the minimum, it should bring together work order costs, maintenance schedules, condition monitoring data and identify flags that may cause problems in the future.
7. The ROI for any predictive maintenance project can range from a few months to a few years, depending on how well the roadmap is built. The first step to any project should always be an analysis of the current state and picking the lowest hanging fruit to generate quick ROI. The ROI generated in the first stages should be reinvested into the process to multiply returns.

Preston Johnson is the Platform Leader for IIoT and Digital Transformation Services at Allied Reliability, with a technical focus on condition monitoring technology and systems. Allied Reliability is a subject matter expert that provides consulting for predictive maintenance. alliedreliability.com

To learn more about how PTC enables smart factories, visit ptc.com

Accelerating the Industrial Internet of Things. IoT ONE is an insights and advisory firm focused on helping companies manage the threats and opportunities presented by the Internet of Things. https://www.iotone.com