What is Maintenance 4.0?
Let’s start with a definition.
Maintenance 4.0 is the application of Industry 4.0 technologies (e.g., industrial analytics, automation, robotics) to Operations and Maintenance (O&M) activities. The business objective is twofold: higher production yield rates by improving equipment uptime and lower operational costs.
In the image below, we depict the current state of Reactive Readiness activities. In this scenario, the industrial plant lacks insights into why machinery fails. As a result, when a failure incident occurs, production must be stopped during unscheduled repair time.
With Maintenance 4.0, industrial analytics are applied to the sensor data that is embedded in plant machinery. Artificial Intelligence is used to identify evolving failure and repairs can be scheduled in advance. During this the period prior to scheduled shutdown, tools and parts can be ordered and repairs scheduled. In addition, Maintenance 4.0 includes the use of automation, robotics and drones to reduce physical inspection and automate repetitive O&M tasks.
Maintenance 4.0: Not Just Technology
A research survey conducted by Presenso and Emory University on the factors that are inhibiting the deployment of Maintenance 4.0 suggest that the primary blockers are (1) a lack of understanding of Industry 4.0, (2) engineering skills shortage and (3) other priorities. These are ranked higher than the budgetary constraints and concerns about the complexity of software and technology.
Further research indicates that the most important factors impacting Maintenance 4.0 scalability are the training of plant employees on new processes and senior management support.
The feasibility of an IoT infrastructure platform or specific technology solution is secondary to organizational support and enabling processes.
Maintenance 4.0 Audit
Assessment is required of the following topics:
Executive Buy-In: Change starts with a C-level commitment. Maintenance 4.0 deployment is costly and the potential impact on revenue and profitability is significant. Executives will need to own the change and allocate resources (both people and budgetary). It is critical that an executive sponsor is both committed and actively engaged on Maintenance 4.0 adoption.
Strategic Vision: The Executive Buy-In addressed above needs to be formalized and the Maintenance 4.0 should align with the organization’s strategic vision. If there is a disconnect between Maintenance 4.0 and the strategic vision, it will likely be designated as an “operational or O&M initiative.” An organization is considered ready if senior management has developed a strategic plan that has been widely distributed.
Organizational Readiness: Given the disruptive nature of digitalization, some level of employee resistance is likely. In the industrial sector, it is common for Operational Technology and Information Technology groups to be siloed. Successful implementation will require the breakdown of organizational barriers and alignment between groups that traditionally worked autonomously. A second issue relates to employee concerns about their ability to adapt their skill sets to new technologies and processes. One potential issue is that current O&M employees will be disadvantaged due to a lack of understanding of Machine Learning. Other employees are concerned that automation and robotics may jeopardize their job security. Industrial plants need to recognize these issues upfront and acknowledge employees’ concerns. Incorporating in re-skilling and training programs in formalized employee development plants should be considered.
Program Ownership: Who owns Maintenance 4.0 adoption? Is Maintenance 4.0 considered an R&D or technology driven initiative? Are internal champions and evangelists assigned? Success will depend on whether Maintenance 4.0 adoption is owned by individuals that are empowered to make change and understand the dynamics and politics of the organization.
Data Governance: Big Data is the fuel of AI driven Industrial Analytics. The first issue that needs to be checked is data accessibility. Is data that is generated by sensors embedded in machinery accessible to industrial analytics applications? Are there technical or organizational challenges limiting access. Does data format, schema and protocol support Machine Learning?
Although the above list is not exhaustive, these non-technical issues should be assessed realistically in order to support successful roll-out.