Running an Initial Deployment (ID) is a critical exercise. It should be treated as a small-scale version of a larger project – in essence, a trial run. Not only will it demonstrate the value a Predictive Maintenance solution can add to your organisation, it also allows you to learn within a controlled environment and time frame. This means you can take the learnings forward to expand the scope of the project across more machines and/or sites.
We have put together five steps to assist in planning for an ID:
- Define success
There are many motivators for pursuing a Predictive Maintenance program. Success can be a reduction in unplanned downtime, reduced maintenance costs, regulatory compliance, improved build quality or one of the many other factors. It is important to identify and clearly define what success means. If you do not have a target, you cannot measure it, the ID will have no direction and will undoubtedly fail.
- What to measure
Once success has been defined, the next step is to identify the key measures required to assess the trial. Typically, this would include both critical machinery and problematic machinery. A priority critical list can be created by looking at historical data: the top 10 failures in the last month and year in the Computerised Maintenance Management System (CMMS) or Enterprise Asset Management (EAM). Then understanding the consequences and risks of failure of each asset.
Problematic assets can be identified through historical maintenance records and downtime reports. These may not be the big pieces of machinery – they could be small parts that have a big impact.
- Management support
A Predictive Maintenance program is not just a maintenance consideration; it is a mindset shift for the entire organisation. Moving the focus from rewarding fixing problems, to celebrating the prevention of problems through active management. An ID can assess the organisations mindset and attitude towards this concept, but management support is essential.
- Make things happen
An ID typically runs for 12 months. During that time the program needs to demonstrate its value. To do this, it needs to have enough to work with. Ideally including a machine failure to predict. With too few machines included, you run the risk of nothing interesting happening. So the Predictive Maintenance solution is not fully tested and its value remains unproven. We advise choosing a selection of broadly similar machines - between 20 and 50. During that time you’re more likely to see something of value.
- Take the organisation on a journey
Change often sparks concern – for jobs, workload, new goals, etc., particularly when it comes to technology in an industrial manufacturing setting.
The project group should be fairly small to maintain flexibility and enable communication. It should include individuals with the necessary skill sets to make the project a success. These individuals will assess and support the program and generally be instrumental in the project success. They will act as the central point of contact for collecting feedback and sharing progress and updates. As well as reviewing of the project both for the initial period and its wider potential, including scalability considerations.
In shortHopefully these tips will assist your preparation for running an ID. But ultimately, no matter how prepared you are, the biggest threat to the success of an Initial Deployment is inaction – failing to get started. Look around the business to put together a dynamic, forward-looking project group and make it happen!