Prognostic maintenance (also known as condition-based maintenance, predictive maintenance, or simply prognostics) is the ability to know the condition of equipment, and to plan and perform maintenance accordingly before a critical failure
A motivation for carrying out regular maintenance is well established in culture. However, the cost of a maintenance activity, or the cost of not undertaking one, are often not clear. Over the last few decades, cloud infrastructure and efficient edge computing have become more and more powerful, and, at the same time, cost effective. These have enabled easy access of deep analytics, machine learning (ML) and Artificial Intelligence (AI) to mass applications. While that is good news, the motivation to adopt any new technology must be much more than the mere availability of it. Adoption of new technologies demands ‘change’, and any change is often expensive and challenging. Manufacturing systems and processes have evolved over decades and have been perfected and time tested. So, what is wrong with the manufacturing systems and maintenance processes in their existing state?
While we try to elaborate the answer to that question in this whitepaper, the simple answer is ‘manually determining when to undertake a maintenance activity is often error prone and not scalable, leading to either too much maintenance or too delayed maintenance, both of them resulting in business loss.’
To The Point – Return on Investment
To drive straight to the point: There is no reason why you should not think of implementing a predictive maintenance system!
A report by Aberdeen Group highlights the cost per hour of machine downtime of $260k! While this may not exactly match your specific use-case but certainly indicates the immense potential for improving business. The various dimensions of downtime cost are complex to determine and would require an in-depth investigation to quantify expenditures and estimate loss around any downtime. There are planned maintenance costs, unforeseen repair costs, human capital losses and follow-on losses associated with your production process, as we will see in the next section.
- Additional spare parts and repair cost
- Higher machine wear and tear and consumption (e.g. fuel)
- Work around and change over efforts
- Unused human capital due to downtime
- Human errors due to error prone case procedures
- Loss due to replacement of parts with significant remaining useful life
- Unforeseen logistics and storage costs
- Production delays & quality impact
- Root cause analysis and mitigation process
- Image and reputation impact for the business in the public