How AI is Revolutionizing Capital Planning and Supply Chain in Health Systems

Summary

Podcast Summary – Becker’s Healthcare Podcast

Episode: How AI is Revolutionizing Capital Planning and Supply Chain in Health Systems
Date: March 31, 2026
Host: Lucas Voss
Guest: Fruleka (Sri Lakha) Akula, Chief Data and AI Officer, Trimedics

---

Episode Overview

This episode explores how artificial intelligence (AI) is transforming capital planning and supply chain management for health systems. Guest Sri Lakha Akula shares industry expertise, breaking down existing challenges, the role of AI in reshaping processes, and the tangible benefits of data-driven, interconnected systems. The discussion highlights why a continuous approach, powered by AI, is vital for today’s healthcare organizations and outlines the risks of inaction.

---

Key Discussion Points & Insights

AI’s Role in Modern Capital Planning

• Traditional Capital Planning Is Outdated:
  • The current process is static, slow, and not responsive to ever-changing needs.
  • “To receive a cogent high fidelity response takes roughly two to three weeks to two to three months actually… It’s very static, it’s a point in time.” (B, 02:20)
• AI Enables Continuous, Dynamic Processes:
  • Shifts capital planning from a once-a-year task to a real-time, scenario-driven, ongoing process.
  • AI lets leaders model different scenarios, make tradeoffs, and update decisions as circumstances evolve.
  • “Capital planning is not a once and done process, it’s a continuously informed process.” (B, 03:37)
• Importance of Scenario-Based Modeling:
  • AI provides the capacity to simulate and plan for device recalls, cyberattacks, or facility changes, offering agility that manual processes lack.

The Data Challenge in AI Adoption

• Fragmentation and Vendor Fatigue:
  • Health systems deal with many vendors and fragmented sources, making it hard to gather comprehensive data.
  • “Healthcare data is one of the most fragmented data systems that we know of in any industry. Not only that, we know that health systems are generally vendor fatigued.” (B, 04:24)
• Necessary Data Types for AI Success:
  • Device intelligence (utilization, remaining useful life, service history)
  • Cybersecurity risks, parts availability, supplier-provided end-of-life data
  • Comparative benchmarking—both individual (“n=1”) and aggregate, across systems.
  • “It’s not only that independent health system’s information, but it’s also an aggregate of temporal data… to be competitive in the landscape today.” (B, 05:29)

From Predictive Insights to Actionable Supply Chain

• Systems Integration Is Lacking:
  • Predictive maintenance and part sourcing often exist in silos.
  • “Hospitals don’t order parts turns out just for the fun of it. Similarly, a machine maintenance is not that complete until the necessary part is being replaced… But the systems are actually built that way, they don’t talk to each other.” (B, 07:01)
• End-to-End Value Chain Enablement:
  • AI bridges predictive failure alerts to real-time part procurement and planning, minimizing disruption and patient impact.
  • Planned downtime (versus break-fix) ensures better patient care and prepares supply chains for just-in-time response.
• Direct Cost Savings and Strategic Value:
  • Improved capital efficiency, minimized revenue loss, and the reframing of medical devices from “cost centers” to “strategic assets.”
  • “These hospital systems, these devices are typically seen as cost centers, but they are actually strategic assets… to make sure that patient care is not interrupted.” (B, 08:42)

Risks of Treating Alerts & Planning in Isolation

• Patient and Revenue Loss:
  • Reactive maintenance (break-fix) can disrupt patient schedules, resulting in rescheduling or patient attrition.
• Supply Chain Waste and Inefficiency:
  • Over-ordering, wrong part procurement, and unnecessary costs accrue without predictive integration.
  • “Procuring too many parts or procuring ahead of time instead of just in time, or having… to return them because it’s not the right part or having to get at a CO costs that are not optimal to the hospital system.” (B, 10:26)
• Delays in Executive Decision-Making:
  • Siloed systems slow down capital planning approvals, justifications, and board confidence.
  • “With AI in capital planning, it makes it easier, faster, better for all of them.” (B, 11:23)

---

Notable Quotes & Memorable Moments

• On the Static Nature of Traditional Capital Planning:

  • “It takes roughly two to three weeks to two to three months actually… It’s very static, it’s a point in time.”
    — Fruleka Akula (02:20)

• AI’s Shift in Process:

  • “What AI allows us to do is to have a dynamic and interactive process that is rooted in data, that is rooted in solid evidence and that makes it more interactive and dynamic and real time.”
    — Fruleka Akula (02:44)

• On Data Challenges:

  • “Healthcare data is one of the most fragmented data systems that we know of in any industry… There are a lot of vendors in a single value chain and that is same for making these capital decisions.”
    — Fruleka Akula (04:24)

• Supply Chain and Maintenance Integration:

  • “Hospitals don’t order parts… for the fun of it… But however, the systems are actually built that way, they don’t talk to each other. And bridging this gap… allows us to bridge that end to end value chain to make these systems failures with minimal disruption and downtime.”
    — Fruleka Akula (07:01)

• On Devices as Strategic Assets:

  • “These hospital systems, these devices are typically seen as cost centers, but they are actually strategic assets for the hospital to make sure that patient care is not interrupted.”
    — Fruleka Akula (08:41)

---

Segment Timestamps

• 00:28 – 01:10: Guest introduction, background, and expertise.
• 01:10 – 03:54: Problems with traditional capital planning, AI’s transformative potential.
• 03:54 – 06:17: Data requirements—types, challenges, and need for aggregation.
• 06:17 – 09:20: Moving from predictive AI maintenance to integrated supply chain action.
• 09:20 – 11:35: Risks and pitfalls when decisions are siloed, impact on patients, and organizational consequences.

---

Final Takeaway

AI is actively enabling health systems to modernize capital planning and supply chain management from reactive, fragmented processes to continuous, data-driven decision-making. Despite the industry's data and vendor challenges, integrated AI solutions promise improved operational efficiency, patient experience, and long-term competitiveness.

---

Guest: Fruleka Akula
Host: Lucas Voss
Brought to you by: Trimedics
For more, visit Becker’s Healthcare Podcast.

Transcript

[A]

Hi everyone, this is Lucas Voss with Beckers Healthcare. Thanks so much for tuning in to the Beckers Healthcare podcast series. It's great to have you. Today we're talking about how AI is revolutionizing capital planning and supply chain in health systems and I'm so excited to have her. Joining me for today's discussion is Fruleka Akula, Chief Data and AI Officer at Trimedics. Fruleka, thank you so much for being here today. It's so great to have you.

[B]

Thank you for having me, Lukas. Happy to be here.

[A]

Absolutely. I do want to start us off with introductions for those that might not know you at least yet. Could you just share a little bit about yourself and your work?

[B]

Happy to. I'm Sri Lakha. I am Chief Data and AI Officer at Trimedics. I lead our product development, data and AI functions here. Prior to that I was Chief Technology Officer for Alto Pharmacy and before that I led countries. One of the largest revenue cycle management software at Optum, including their AI disruption and prior life includes work at Google and Amazon of the world and FinTech etc.

[A]

I said that it's so great to have you and that's. I mean that in the truest sense because you have bring such a vast knowledge about AI and about the space with you. And I do want to start us off with sort of laying the basics for our conversation today. Right. What is AI doing differently in capital planning today and how should leaders think about its role alongside human decision making?

[B]

Yeah, it's interesting, right? Overall, if we look at the Gartner's hype cycle on when first AI appeared as an intelligent agent, it's been roughly over three decades since it has happened. And AI in healthcare now, and very specifically within capital planning, if we take stock of that. A recent study in HFMA shows that the average age of a plant in a not for profit hospitals is roughly about 12.8 years in FY24. That's the oldest average in at least the last 13 years. And not surprisingly that I showed a surge in capital spending. Right. And within that capital spending today, if you take a step back and look at how the capital planning cycle happens, it's pretty static from the point where a CFO or a capital planning director asked the question about hey, I have XYZ budget and where do I spend it? To receive a cogent high fidelity response takes roughly two to three weeks to two to three months actually. And God forbid if they have additional questions on top of that, it's a few more weeks on Top of that, right? It's very static, it's a point in time. And what AI allows us to do is to have a dynamic and interactive process that is rooted in data, that is rooted in solid evidence and that makes it more interactive and dynamic and real time. And also allows for scenario based modeling, which is hard to do in today's situation where if you were to make judgment calls or trade offs between different decisions and different scenarios, it is easy to make it with AI in healthcare now than what it used to be. And I'll also add this Lucas is capital planning is not a once and done process, it's a continuously informed process. For example, a device could be recalled, a cybersecurity attack could have happened which needed a system to take down or new facilities are being added all the time. So capital planning is not once in a once and done, once a year process, but it's actually a continuously informed process. And the AI disruption in healthcare now allows, and especially in capital planning allows for us to have that continuously informed process rather than a static point in time process.

[A]

I love that you mentioned that because I think when it comes to that, to be able to have that continuous process, what you do need is also continuous data. You need data that fuels that process to be continuous, which is so crucial. As you've just outlined, what kinds of data does AI need to make better capital decisions? And why is that still so hard for health systems to pull together?

[B]

It is hard. Healthcare data is one of the most fragmented data systems that we know of in any industry. Not only that, we know that health systems are generally vendor fatigued. There are a lot of vendors in a single value chain and that is same for making these capital decisions. To make an informed and well educated, deeply rooted in data and evidence decisions for capital planning, it needs the data from device intelligence, the utilization data, the device age remaining useful, life or the end of life as prescribed by the suppliers, or the service history, cybersecurity risks, parts availability, all of these play into making very informed decisions on capital planning. And that gives you the information about how and where a particular hospital needs to make investments in based on their use. That's n equal to 1. However, other thing that would be missing in that case would be an aggregate. How does that particular health system compare against its likeness of other health systems? That information would still be missing. So it's not only that independent health systems information, but it's also an aggregate of temporal data that is taken over decades of information that makes it the most comprehensive, informed decision for a health system to not only do it at n equal to one individual unique insights into their capital planning, but also compare them against the rest of the ecosystem to say how well they're doing and how can they be continued to be competitive in the landscape today? It's poor visibility, it's fragmented, it's incomplete data. All of those contribute to the difficulty in fetching this information.

[A]

Yeah, we come back to what you've mentioned earlier, right. It needs to be a continuous process. And this what you've just described, if that's not done right, that continuous process can't exist. It doesn't work for organizations that way. And I want to turn this into action a little bit and sort of discuss the scenario here. You've touched on the multi vendor piece here too, which is certainly crucial. Once AI predicts likely failure. Right. How should that insight then translate into real supply chain action? And where do health systems typically see breakdowns in some of this? Connecting predictive maintenance with that multi vendor part sourcing piece?

[B]

Yeah, these are fragmented systems as we've been talking about. Hospitals don't order parts turns out just for the fun of it. Similarly, a machine maintenance is not that complete until the necessary part is being replaced and it's fully repaired. It sounds silly when I say that. Right? But however, the systems are actually built that way, they don't talk to each other. And bridging this gap between the two systems in this context of predictive maintenance and the failures and connecting that to the supply chain allows us to bridge that end to end value chain to make these systems failures with minimal disruption and downtime. A planned minimal downtime implies eliminating patient impact at the end of the day. That's what we want is minimal, if at all patient impact. And that's what it allows us to do by combining these two systems and informing one system to the other. The reason it is important for us to take stock of multi vendor intelligence of these parts is to allow for those predictive failures when those happen and when they are predicted. Well, using AI, being able to look at what is the best place and best way for us to fetch a part to replace and maintain that system is important. It allows for hospital systems to be capital efficient and it allows for those direct savings to be translated back into the health system to repair it. Again, going back to the patient experience so that the planned downtime happens. It's not just a break fix at that time responding reactively, but it's actually planned and it is procured before the the failure happened, ready to go and replaced and that machine is repaired. And it's important that these two systems is connected to make sure that actually happens. And at the end of the day, the way it translates to is it's direct savings from plant supply chain, it's also direct impact on the patient experience and it's preventing loss of revenue. Right. These, these hospital systems, these devices are typically seen as cost centers, but they are actually strategic assets for the hospital to make sure that patient care is not interrupted. And it's important as a result to make sure that these are not only predicted, but also connected all the way through to supply chain and procured ahead of time to minimize that and to prevent that loss in revenue.

[A]

And I want to touch a little bit more on that too because obviously loss of revenue is something that you've just highlighted that's really, really crucial to what are some of the other risks that you're seeing when all of this is treated in isolation, when some of these alerts are treated in is of this protective predictive maintenance that we've touched on is treated in isolation. Again, without tying all of this to inventory availability, costs, capital planning, et cetera, what's the risk apart from just the revenue loss that you just described when we're treating all of this in isolation?

[B]

Yeah, let's walk through a scenario, right? If these systems are not connected, what is the risk that we see operationally it is the system's maintenance is, is a break fix. At that time the system needs to be taken down, the patients need to be rescheduled and that could actually lead to a patient loss for that hospital system. Because you know, they, they don't want to sit around waiting for this to come back up or to be rescheduled. And that's a real loss to the hospitals. And all the way through to inefficient supply chain systems is another risk is procuring too many parts or procuring ahead of time instead of just in time, or having the having to return them because it's not the right part or having to get at a CO costs that are not optimal to the hospital system. These are all the various different risks that we see when we are looking at not having a strategic multi vendor strategy at a hospital system or when these predictive maintenance systems are not connected to proper sourcing and supply chain systems. An efficient supply chain process presents a solid case and justification for the hospital systems and for the capital planning directors and the CFOs and the CMOs and the CIOs to get a buy in from their executives collectively and from their board to make these decisions faster. At the end of the day, the speed to making decisions and having the confidence in the decision that they are making and give the sufficient justification to their board is one of the most hardest things that they'd have to do year over year. And with AI in capital planning, it makes it easier, faster, better for all of them.

[A]

Trulika again, thank you so much for your time and insights today. We also want to thank our podcast sponsor, Tremedex. You can tune into more podcasts from Becker's Healthcare by visiting our podcast page at beckershospitalreview. Com.