Risk is the ultimate measuring stick for gaining the biggest impact, and letting the important assets speak loudly to our investment plans. Determining risk across all asset types sounds like it could be a daunting process. But, we know what is important to our organization, and we know a lot about our assets, so if we combine it with a little more “real world” analysis, we can find a clear way forward.
Risk, simply put, is the product of the probability of an asset failing and the impact of that failure, less any mitigation strategies that we may have in place.
Probability of Failure (POF) and Failure Modes
Assets fail, but they can fail in different ways, at different stages of their lives, for different reasons. The trick is to make a prediction of failure far enough in advance that we can actually plan for it. The simplest way of determining the POF of an asset is to flip the condition curve upside down, and watch for the asset to hit the bottom of the curve. A method that has more usefulness as a key to managing risk is to look at failure modes.
The first step in figuring out the POF of an asset is to sit down with the people who actually look after them and list out all the reasons they need to replace those types of assets. Do they get hit by cars, fall down on their own, use too much power, cost too much to fix again?
There are four key modes of failure:
- A failure of capacity, like a distribution network that cannot cope with demand.
- A failure in the level of service, when an asset fails to deliver the minimum acceptable customer experience.
- A failure due to economic efficiency, like an asset that costs more to operate than it does to replace.
- Finally, and most common, physical mortality, when the asset simply ceases to function.
Consequence of Failure (COF)
Once we determine how an asset can fail, and how likely it is at any given time, we must determine just how much impact that failure will have on our organization. The organizational objectives and overall asset management strategy are a great reference when we start to try and quantify that impact.
Typically, we start by looking at an asset’s failure impact by creating several scales, and ranking the impact of a failure on those scales. A typical approach is to look at social, economic and environmental impact of failures, but your COF scales can take whatever form is important to your organization.
The critical thing is that the scales are relative – so that they can be applied across the asset inventory. This is the key step – allowing us to compare apples and oranges, wires and transformers, substations and properties, all on the same scale.
Ranking by Risk
Once we have POF, and COF, we can combine those to come up with a risk priority. Each asset can then be ranked according to the exposure that the organization would experience if the asset failed, based on multiple causes of failure. The end result is a risk priority number (RPN) value for each asset in the inventory.
This value can be used to drive inspection frequency, insurance valuations, environmental mitigation strategies, and other regulatory activities, and also rank competing events and activities.
By having a clear grasp on our asset data, we gain enormous clarity in understanding the scope of our activities. Not only does our reporting become simpler, but we are able to quickly answer questions about current valuation, condition and risk exposure. This flows automatically into our updated asset management plans. The result is we can separate out the most critical assets from the pack, and act on them first.
Building Asset Strategies
Knowing what we can do is critical to managing asset performance. Knowing whenand when not to do it is just as critical. Managing asset inventories is the art of balancing operating and capital costs, determining the best point on an asset’s life cycle to inject activities or interventions to keep the costs over the asset’s life at their minimum, while maintaining the value of that asset to the organization.
Capturing Asset Behaviours
The key to developing these strategies is to work directly with the asset managers, and extract the institutional knowledge they have about how those asset actually perform, and what we can do to them in order to keep them performing in the real world.
That does two key things for us: it gives us a great start in building a decision tree to predict future investment requirements, and it captures that institutional knowledge on how best to maintain those assets.
Once we’ve documented it, it will serve as the starting point for new staff as they come on-board, shortening their ramp-up time dramatically. The resulting decision logic, when applied to individual assets, generates a lifecycle forecast of what can be done to the asset over the course of its life, complete with costing, risk impacts, and value contributions.
The product of this is the list of pending asset events and activities, based on everything we know about our asset inventory, and using our own institutional knowledge.
This can then flow into the defined asset model and be used to generate ongoing forecasts of investment requirements.
Each asset contains all of the attributes defined for it, such as age, in-service, material, diameter, voltage, etc. It also contains a set of events and activities that have been generated for that asset based on our strategy, predicted values of future measures, and probability, consequence and risk profiles.
By doing it this way, we can make continual improvements to the asset-type definition and strategy, and apply it to our asset base to get better predictions and plans as we move into the future.
We can also set up alternative strategies, failure modes, and decision models to play what-if scenarios with our asset base. The key output of all of this is our needs list, for next year, the year after, or 10 to 50 years out.
Elements of the Asset Management Plan
All of the components of the asset management plan are now in our hands, and can be used to generate the plan at any time. These components include the textual entries that describe the current state of the asset group populating the plan, as well as the raw data that populates the tabular and graphical elements of the final plans.
A typical asset management plan should include:
- Asset base (inventory counts by asset sub-type)
- Asset failure modes (failure modes for each asset type within the group)
- Asset costs (unit costs for replacement, maintenance and monitoring)
- Projected inventory (graphical representation of asset counts for next 10 years)
- Age profile (graphical representation of asset age by year)
- Consumption profile (graphical representation of the percentage of asset life consumed)
- Health profile (graphical representation of asset base condition)
- Maintenance program (graphical representation of asset OPEX projections for next 10 years)
- Planned replacements (graphical representation of asset replacement events for next 10 years)
- Future cashflow (graphical representation of OPEX and CAPEX investments for the next 10 years).
So we have concluded our journey in developing asset management plans using enterprise asset management best practices. We’ve seen how defining organizational priorities help drive asset inventory improvement, and ultimately the format and function of the planning process itself. We’ve also seen that EAM without actual asset data is like paint-by-numbers without the numbers – pretty random! Finally, using management best practices to determine asset priorities is critical to getting the most out of every precious dollar!