Capital planning for large infrastructure portfolios such as airports, university campuses and hospitals is a complex and resource-intensive endeavor. With thousands of assets to manage, facilities teams constantly face challenges. They have to prioritize repairs, replacements, and investments in a way that maximizes operational efficiency and minimizes long-term costs. One of the most critical steps in this process is identifying which assets are the most problematic or “bad actors”. It’s also important to understand which ongoing problems can be classified as ‘defects’ that lead to reoccurring work.

Bad actors are assets that consistently generate issues, require frequent maintenance, and accumulate long work order histories that are costly to the organization. These assets may impact performance, system uptime, and incur costly replacement of parts. They may also influence asset lifespan, and overall maintenance productivity. The result can be significant ongoing costs and productivity drain. Past experience shows that defect elimination alone can reduce downtime by up to 90% and operational costs up to 60%. An estimated 4% to 12%” is reasonable to expect given a solid maintenance program with defect elevation strategies in place.

Traditionally, capital assessments are conducted every few years rather than continuously and in real time. Assessments may be conducted monthly or annually. More commonly, they’re conducted every 3-5 years for end of life assessment of a large asset group. What’s often missing is a view of equipment health over time. As a result, bad actors go under-reported and oftentimes unrecognized by planning departments as maintenance teams battle to keep operations going in a reactive mode of daily activities.

For equipment that has telemetry data and work order history, continuous asset health monitoring can be trended over long periods of time. This helps calculate asset life and compare cost of repairs versus cost to replace. Historically, the process would be a time-consuming data-heavy exercise, requiring deep dives into disparate systems, manual analysis of work order logs, and cross-functional collaboration to piece together a clear picture of asset performance. Today, this process has been simplified with Willow Copilot.   

With a carefully constructed prompt, Willow Copilot can quickly surface assets with recurring issues, long resolution times, and high maintenance frequency. These insights help organizations make informed decisions about repair vs. replace. AI does the heavy lift, adding unprecedented levels of speed to the capital planning process. The system flags assets with frequent failures and long resolution times, then adds context for further investigation. The Knowledge Graph provides a deeper understanding of asset types, locations, components, spaces served and energy impact. This enables decisions backed by data and insights. Additionally, Willow Copilot can query telemetry values to report and calculate equipment state and potential degradation over time. This drives individual monitoring of each piece of equipment resulting in a faster, smarter, and more transparent process.  

How DFW Tackles the Problem with Willow Copilot

Dallas Fort Worth Airport (DFW) exemplifies this challenge with its scale and complexity. This makes it essential for a digital twin and AI powered solution to understand asset health and drive effective operational decisions. Across terminals, non-terminal buildings, parking garages and runways, DFW hosts a vast network of critical systems. These range from HVAC, MEPs, baggage handling, to conveyance (elevators, escalators), lighting, and more. Each system contains multiple assets and components. Work orders and preventative maintenance records are housed in Computerized Maintenance Management Software (CMMS) applications. This includes parts, schedules, invoicing, and much more. Willow ingests CMMS data and telemetry data from existing systems to bridge together a complete understanding of asset health and performance over time. Willow’s AI skills perform complex monitoring, which drives predictive and proactive insights for review and creation of work orders directly from the Willow platform.  

Turning insights into asset health relies heavily on the quality of work order data. Specific details of the problem, recommended actions, problem codes, and resolution comments are particularly impactful. With the right set of prompts, Willow Copilot provides context on recurring issues, root causes, and effectiveness of fixes. When detailed and consistent, these prompts empower AI-driven analytics to identify problematic assets and streamline maintenance planning priorities. This enables smarter capital planning across DFW’s complex infrastructure. 

Kelly Watt, Program Manager for Digital Twins at Dallas Fort Worth International Airport (DFW), shares how Generative AI in Willow Copilot is simplifying this complex journey. 

The system identifies problematic assets using the following criteria:

1. Number of work orders over a period of time 

2. Number of insights over a period of time 

3. Severity of these issues 

4. Is the trend increasing or decreasing 

5. Who is the Maintenance Technician performing the work 

Let’s walk through a sample prompt.  

Disclaimer: Responses may vary. Data points listed are examples only. Energy and cost savings reported by Willow Copilot are dependent on the number of insights enabled in Willow.  

Willow Copilot helps identify top problematic assets as well as opportunities to improve data quality. This has led the team at DFW to implement new processes that now require technicians to enter detailed closing comments in work orders. In addition, the “Solution” field is being templatized to ensure that key information is consistently captured for better analytics and decision-making. 

Conclusion 

AI and Willow are transforming capital planning by making it faster to do the heavy lift of contextualizing work order history against critical events on high value assets. This creates a new dimension to Willow’s digital twin, combining real-time insights with asset health to enable smarter planning. With these tools, facilities teams can connect actionable insights to the work that resolved them, realize avoidable costs, spotlight the bad actors amongst all assets, and transform capital planning into a more efficient process.