While enterprise asset management (EAM) can be necessarily reactive, it ideally includes a strong strategic component. CFOs are inundated with capital requests. How do they decide which to pursue and, ultimately, answer the question: What is my organization’s best allocation of precious capital?
EAM has the power to shape decisions at the top, but it needs to be informed by data from the front line regarding an organization’s assets throughout their life cycle. Strategic EAM can drive planning, executing and tracking of maintenance activities across departments and locations to optimize return on investment. If done wisely, EAM can promote significant savings.
For example, one industrial business regularly purchased a certain piece of equipment with a seven-year life cycle. However, data showed that in practice this equipment consistently performed at a high level long beyond the expected seven-year time period. It continued to perform well and generate impressive cash flow. By using this data, the company could responsibly allocate capital to other uses while avoiding premature purchases.
Using EAM leads to smarter decision-making. While it may be necessary to plan for equipment replacement, it’s equally important to optimize an asset’s actual useful life.
Similarly, data could predict when equipment needed to be replaced, even if that result was sooner than expected — perhaps because the equipment was stressed in an unforeseen way — thus avoiding expensive outages.
In short, strong asset management is risk management, which means it’s opportunity management, too.
The chain of insight
How do you get to the point where you can make this kind of case?
The insights outlined here derived from business judgment applied to predictive analytics, which employs data visualization, data mining and artificial intelligence (AI). Predictive analytics tools are fed by a robust asset inventory; asset inventories are facilitated by asset management systems.
You’ll want to keep these two considerations in mind:
1. Predictive analytics considerations
Predictive analytics applies algorithms to data models. Ideally, these models are built on multiple data sources. The raw data usually has to be cleaned, possibly integrated, and combined with other data sources to derive new data variables.
AI can be used to process unstructured data, such as verbatim reports, and to process overwhelming data, such as archived inspections checklists.
Fortunately, AI — in the form of automated cognition — is uniquely capable of managing different types of data (text, speech, photos, e-mails, etc.) and provides the ability to query all of that data without a pre-built schema. For example, AI can be used to process unstructured data such as verbatim reports. It can also be used to efficiently process formatted but overwhelming data such as archived inspection checklists.
Once you’ve arrived at a sound data model, various statistical, data-mining and machine-learning algorithms are available. Select the algorithm(s) based on the objectives of the model and the data to be analyzed. It’s important that the variables selected have predictive powers and can be tied to business objectives. Some predictive analytic projects succeed best by building an ensemble model, or a group of models, that operate on the same data.
2. Asset inventory considerations
This inventory is generated by a unified system for asset management. In some cases, organizations must improve business processes and upgrade IT infrastructure – this can be done incrementally.
Data visualization can transform the complex into easily understood visuals. Predictive analytics can sort data into information and information into actionable priorities.
Establishing a best-in-breed asset management program means double-checking formulae, grouping assets into portfolios, establishing audit trails, and applying current depreciation methods and relevant regulatory standards. Data is gathered via real-time visual inspections of properly tagged assets (ideally with hand-held devices by people aware of relevant performance standards); subject-matter-expert and operator input; historical records; and tangible resources such as physical plans, asset schematics and “ghost inventories” of lost or missing assets.
Toward a more strategic role
Once you’ve applied relevant algorithms to this rich data, you can step into a more strategic role. How can you best communicate your findings to the C-Suite? How do you get their attention? Inform their decisions?
The key is expressing your insights in their language. This means thinking in terms of time and the time-value of money.
In presenting to CFOs, it’s often useful to frame findings around three fundamental financial statements:
- Balance sheet
- Income statements
- Cash-flow statements
Every insight should be framed in terms of how it improves fundamental financial performance over time.
In addition, the ability of data visualization techniques to render complex data into easily understood visuals is a powerful tool. Similarly invaluable is the ability of predictive analytics, framed by your experience, to sort data into information and information into actionable priorities.
Ultimately, asset management will help you add new kinds of value to your organization at the highest levels.
Grant Thornton Principal
EAM practice lead
+1 617 848 5060
Grant Thornton Senior Manager
+1 312 602 8015
Grant Thornton Director
+1 919 881 2718