Condition-based (CBM) or predictive maintenance (PdM) endeavours – Proof of Concept, Proof of Value – are often uncharted territory for organisations. This is either because data is unlocked from assets where no-data has been before or because the organisation is going to trial new methods of service/product delivery where no other organisation has gone before. This all sounds a bit tricky, but deep down the innovative nature of many IIoT endeavours can also be havoc to its result.
This post describes key learnings that we have experienced at UReason in various CBM/PdM endeavours; they are centered around:
- Leadership & Support
Starting your CBM/PdM journey without defining a clear process can be disastrous to the results. Minimum ingredients for a successful process are clear business requirements and evaluation criteria. Next to this try to limit the start-to-end of the PoC/PoV to 8-12 weeks.
Depending on the nature of the project, define a number of feedback points in the project based on results that you achieved. These can be the integration of systems, evaluation of technology/products, demonstrations that have been readied et cetera.
In summary: run the mini project with a clear process.
The number one failure trap for your CBM/PdM PoC/PoV is growing ambitions. Successes in the execution of your project can lead to scope creep which inevitably leads to a disruption of the process, extension of the execution time and often to a decrease in support.
Even when you have defined clear business goals for your PoC/PoV be sure that you do not over ask the team/organization. Fence the ambition of your PoC/PoV by keeping a close eye on the evaluation criteria. If a newly found requirement is already tested or evaluated in a different form, as part of your project, ask yourselves why adding it will make a difference. In summary: fence your ambition!
An excellent way to prioritise the requirements (existing and new) in your project is to apply the MoSCoW method:
|Must Have||Requirements that are critical to the success of the PoC/PoV. These are typically the minimal requirements that the project must complete in order for it to be a success.|
|Should Have||Important requirements, but not critical to the success of the PoC/PoV. Keeping a tally of these during the project pays off when the PoC/PoV is successful as they often contain the best next items to consider for a wider rollout of your CBM/PdM project.|
|Could Have||These requirements are often nice to have but not a necessity for the success of your project. As with the Should Have requirements, keep a tally of them as they may come in handy in later post PoC/PoV stages.|
|Won’t Have||Sometimes this subset is called ‘Would Like to Have’. They are really not a necessity to the project, what is important for these requirements is to document why you don’t want to have them in your project.|
The rationale for assigning requirements a must, should, could or won’t-have priority can either be based on group input or on a method that calculates the importance of an individual requirement based on its criticality, value-add et cetera. When the time-line for the delivery of your PoC/PoV is clear, the rationale for assigning requirements sometimes is the time-limitation at hand. In such cases clearly document why a requirement did not make the deliverable!
Leadership & Support
Due to the innovative nature of many of the CBM/PdM Poc/PoV endeavours leadership is often not the issue. A strong leading champion is often present as they are (often) the initiator. The issue we find frequently is in the support of the organisation for the sometimes somewhat game-changing endeavours.
There is no cookbook on how to obtain support from existing organisations and groups. It is all down to clear and regular communication. Clarity as to why the PoC/PoV is run, but also clarity about what it can mean for the organisation when successful. The leader of the PoC/PoV must also be the ambassador, special envoy, of the initiative within its own organisation to obtain optimal support.
UReason has been at the forefront of condition-based and predictive maintenance, reasoning over real-time streaming data and events in the manufacturing industry and telecom. We apply an ensemble of techniques – best fitting the requirements – and a wealth of knowledge focused on providing a tailored response to the environment of our customers.
Our capabilities in the maintenance field include:
- Feasibility studies and Proof of Concepts including hardware prototyping and field tests;
- Support and roll-out of IIoT solutions in Operational Safety and Predictive Maintenance;
- Recommendations for human-cyber physical systems, augmented reality and Internet of Things technologies; and
- Support in Machine Learning and Big Data initiatives supporting IIoT applications.