Automating Warehouse Inventory with Drones: An MIT SCM Capstone Project

- [Mackenzie] Welcome to another episode of "Supply Chain Frontiers," the MIT CTL podcast where we explore the trends, technologies, and innovations shaping the future of supply chain management. I'm your host, Mackenzie Berry. In today's episode, we're venturing into a cornerstone of the MIT Supply Chain Management Master's program, which is the capstone project that students complete with a company every year. Here to join us are Tommaso Portaluri, Camilo Mora, and Elisa Ruiz.

- [Camilo] Hi, everyone. My name is Camilo Mora. I'm a Postdoctoral Associate at CTL. And, well, I was invited to participate in this project a year ago. I joined Dr. Josue Velasquez in advising very talented students to develop this project aimed at, you know, better understanding the potential improvements in terms of reductions of scope to emissions by replacing forklifts with drones for, you know, warehousing activities.

- [Elisa] Hi, I am Elisa Ruiz. I was part of the Supply Chain Master program last year, Class of 2025. And when I was there, I chose this project. This was my first choice as a capstone project, and it was given to me, fortunately. So I had the opportunity to work with Tommaso, Camilo, Dr. Josue and Dr. Miguel. And I was interested because I come from a work background in warehouse operation and automations, but I have never looked at it from a sustainability lens. So I thought it was very interesting to have that opportunity to know, like, the impacts it might or might not have in sustainability.

- [Tommaso] Hello, everybody. My name is Tommaso, and I lead the sustainability efforts at Verity. Verity is a Zurich-based company with global operations that has developed a system of AI-powered drones that automate inventory in warehouses. What was very interesting for us was the fantastic work that the Sustainable Supply Chain Lab has been doing in the last two years within the MIT Center for Transportation & Logistics. This capstone was the first opportunity we had to actually try to put some hard numbers on the sustainability benefits that our solution enables in one of our client's warehouses.

- [Mackenzie] Glad to you have you. Tommaso, you spoke to it a bit, but I wonder if you could speak a bit more to what inspired Verity to work on this particular capstone project and what you all were hoping to learn from it.

- [Tommaso] Absolutely. We do have quite a strong operational story of what are the benefits that our solution brings in terms of accuracy, labor savings, error scanning and so on, things that are relatively easy to measure. But we're really lacking a rigorous and independent quantification on the environmental benefits across all the three scope of greenhouse gas emission. And so what we're really looking for in this project is to work with academic rigor and to partner up with our clients and with MIT and try to understand better what are the benefits that we're enabling for our clients.

- [Mackenzie] And what drew you to focus on the environmental impact of warehouse automation, as opposed to just focusing on operational efficiency over anything else?

- [Tommaso] I would say there are two aspects. The first one is the operational efficiency was already well proven, so there was not much to discover there. And the second one, that despite these two being often presented as a trade off, what we believe in is that whenever you're going in the direction of sustainability, you're also improving operational efficiency. Quite interesting in the last year, the Sustainable Supply Chain Lab also compiled a report that I think was presented in a previous episode, and operational efficiency was named there as the number one near-term decarbonization driver. And so for us, it was really important to look quite at environmental impact but also look at how they interact with the operational efficiency and try to limit a business false meet of the trade-off between these two components.

- [Mackenzie] And Elisa, you spoke to how you were working on this as a capstone and you spoke about what drew you in terms of the environmental angle. But wonder if you could speak a bit more to what you were interested in about the project and maybe even how you were looking at it for your career at that point.

- [Elisa] Yes. So, as I mentioned before, I was working with warehouse automation before going to MIT, and it was always very focused on increasing capacity, taking more advantage of automation to improve efficiency, throughput, and very operational focused, but I never, ever was asked the question of what, like, every effort we were doing at the time, how was it impacting the environment, right? So, when this project was presented to us, I thought that was very interesting also from the inventory management perspective. Because inventory management, like, when you think about warehouses and operations and the manufacturing operations, no one really cares about inventory management. So, like, most of the time, it's very overlooked, and warehouse sustainability is another area that is also very much overlooked. So I thought about, "Okay, this is a good chance to kind of like bring the spotlight to those two areas and see how they together can create something interesting for the environment." So, after working on this project, I was not thinking about technology or the automation alone, but I was thinking about how information quality affects sustainability. So it's not only about how we are gonna automate all the processes and now use all of these machines, but how those machines can transform the process in a way that it's affecting everything around them, including the environment.

- [Mackenzie] And, Camilo, can you speak to, from a research perspective, why is it so important to quantify the sustainability impact of drone-based inventory systems?

- [Camilo] From a research perspective. Not just about this project but others, you know, it's very important to really assess with, you know, objective lens whether they are, in this case, greener or not, right? Whether they are more, you know, efficient or not. And with, like, you know, again, from an unbiased standpoint provide, you know, evidence to determine whether this is true or not, right? So, now in terms of warehouse automation, it's often evaluated mainly through an operational lens, right? And then the question is like, "Does this technology improve productivity or reduce labor requirements or, I don't know, increase inventory accuracy? So, again, those are, like, important questions, but in this case, they do not really capture the sustainability implications of the technology. So if companies are going to claim, you know, in this case that autonomous drones can contribute to decarbonization or environmental performance, we need to move beyond intuition and quantify those effects rigorously than having to partner, in this case, Verity, sponsoring this project to understand greenhouse gas emissions across the scope 1, 2, and 3, and then, you know, making a more deeply, you know, focusing on scope 2 emissions with the second project and build against something that was not just evidence-based, but, I would say, with scalable recommendations for other companies in the future.

- [Mackenzie] Before we dive into the findings of the capstone project, for those who may not be familiar with Verity's technology, Tommaso, can you explain what autonomous drone-based inventory automation actually does in a warehouse?

- [Tommaso] So, Verity developed a system of a swarm of autonomous drones. So, nobody is piloting these drones. They navigate autonomously the warehouse, also high beads even in total darkness. And what they do, they go to the different locations and they scan the barcodes or they count the boxes and the items that are there, and then they produce a report, highlighting discrepancies, misplaced pallets, lost pallets and so on. And they're able to do this with very high frequency, much more that would be possible without automation, just with the the human-based scanning. There is no physical infrastructure chain to the building. And this continuous update of the app to really allow to have a digital twin of the warehouse, where you have a very high visibility on what you have in stock or what you don't have and what are the problems that they materialize before they materialize the worst model possible. It is usually when you need the goods to be shipped to some customers.

- [Mackenzie] And how does this differ from other forms of warehouse automation that we've seen and why is the focus on inventory counting specifically important?

- [Tommaso] So, if you think of many automation solutions that we see nowadays, logistic fairs or around conveyor citation, ISRS or forklifts, it's mostly about moving goods. For us, what we really wanted to focus on was the data visibility layer or, as I said before, is the information quality. And so we don't lift anything, but we generate evidence, we provide visibility to our customers, you know, where the goods are and how they can intervene to reduce inaccuracies. And once you operate at this level, the ripple effect of increasing by 1%, by 2% or by 5% your inventory accuracy really has a multiplying effect then on your picking activities, on your shipping activities and so on. I think this is why it was so important with this project to really look at the impact on the entire value chain and across all emission scopes.

- [Mackenzie] And Elisa you were speaking to this earlier, why do you think inventory management is underrepresented or under prioritized in the space?

- [Elisa] I think because most of the times, it's only treated as a metric. As a KPI, you do your cycle counts. You say, "Oh, we were supposed to have 100 items, I only found 99, and let's write off one item and that's it, right?" But no one ever really gives attention to all the consequences that one lost item can have downstream and upstream as well. So, it's not often seen as a process that adds value when we're talking about a whole manufacturing process, so it's often not prioritized.

- [Mackenzie] Pivoting now into the findings from the capstone project, your research found a 49.5% reduction in total emissions from drone implementation, which is quite significant. Can you all walk us through where those emissions reductions actually come from and what surprised you most?

- [Elisa] Yes. So this is actually a funny story because when we came into the project, we thought it was gonna be something very straightforward, very easy. You're changing a process that used to be done by people, now it's gonna be done by drones. And you think, okay, energy is gonna change, the process is gonna change, we're no longer gonna have people and that's gonna be our emission savings. Easy, right? But then we found out that not only the equipment itself, okay, drones, if you compare a drone with a forklift, it's tiny, and also the energy it consumes or the energy source differs. It comes from a battery, whereas forklifts can use fuel or also batteries and energy as well. So the change in terms of, in regards to equipment, it was not very significant. Like, the footprint was tiny, but also the forklifts as a whole is not very polluting either and that was surprising. We thought that a forklift, because of all the components and manufacturing process and its size was gonna be more significant. But not only that, when we were digging more into how the operations were done in the warehouses, we talked to people, we interviewed people, the actual operators that were doing the job, we realized that the process didn't fully shift from still needing manual operations. So, it was not like I used to have people doing this and now I got rid of everyone and I'm only using drones. People were still needed and they were still performing things like looking for items that were lost or accounting for discrepancies, for example. So then, when we noticed that it was not, okay, I got rid of four forklifts and instead I'm using three drones and then I can say those were my savings, we had to dig deeper because we didn't have results, right? So it's like, okay, now where are these savings actually coming from? What the drone changed, right? So by this time, our whole capstone kind of like took a turn because then the technology was no longer the story as we thought in the beginning. But we realized that the operational changes enabled by the technology were actually the story. So in the end, what we found, like, this 49.5% did come from energy, forklift energy. There was a lot of savings there because drones, they require less energy and they're more efficient. It also came from labor, commuting and waste. And that was our most, I think, shocking finding that waste was a big contributor because of what I mentioned earlier of all of these efforts being done into, like, finding an item that was lost, and then when the item is not found, it has to be considered waste because of all of the upstream and downstream emissions that it comes with losing an item.

- [Mackenzie] Absolutely. So really looking at the picture, the operations as a whole as opposed to just isolating each part. And one of your key findings is that inventory write-offs or scope 3 wastes account for about 40% of post-implementation emissions. You spoke to it, but why? Why is waste reduction such a powerful lever when you're looking at warehouse sustainability?

- [Elisa] Yes. So if you think about your everyday products, where are they coming from? They needed raw materials to make it, a manufacturing process and a shipping process, transportation to get to where they need to be, right? So the product is already done, it's sitting at a warehouse and then it gets lost. So, that product is now unaccounted for and those embedded emissions are now wasted. So, it's not only the disposal if it's later found that's generating waste and emissions, but it's all of those things that were done beforehand to get the product to where it was that now was wasted as well. Also, often the company still needs that product, so you're trying to find your product either to ship it somewhere else or to use it for another manufacturing process and it's not there. And it needs to get replaced because you still need that product. So another cycle begins. You need to produce that product again and there comes more emissions. So waste, when you lose a product is not only disposal but is replacement and all of the emissions that came with the product for it to even exist. And that's what makes waste a big, big player into all of these emissions contributions.

- [Mackenzie] And you mentioned employee commuting and that emerges as the largest emissions contributor post-implementation at 37%. Most people think about energy use first, as you mentioned. What does that suggest about warehouse sustainability strategies overall when you're zooming out, looking big picture?

- [Elisa] Yeah, so that could be a little misleading if it's not interpreted correctly. It became in largest after implementation doesn't mean that it got worse, it only means that everything else got better. So because the system improved and commute is, or people, it plateaus, right? You reach a point where you can no longer not use people in your process because you still need to do all of these manual things as to go find some items that were lost, for example, analyzing data, improving operations. You need people for that. So as the system continues to improve inventory, inventory accuracy improves, data improves, then you still have people and that becomes now the biggest contributor post-implementation. So, the lesson is not that commuting gets worse, for example, but it's that sustainability priorities shift as the operations improve and also the use for the workforce, right? So before, you had people doing all this manual work, riding the forklifts, counting manually item by item, and now you have them doing other tasks that probably could add more value.

- [Mackenzie] So that's a good problem to have, that everything else was improved so much that it appears larger than it is. And you all use the GHG protocol framework to measure scope 1, 2, and 3 emissions. We've talked about that in previous episodes of the podcast. But for you all, why was it important to look across all three scopes and what did taking that broader view reveal that maybe a narrower focus might have missed?

- [Camilo] Sure. The first study looked broadly at greenhouse gas emissions across, as you said, scope 1, 2, and 3, and showed that drone-based inventory automation can reduce emissions, right, through several channels, including, you know, as previously described, including lower forklift use and reducing inventory write-offs, you know, amongst others. And then, in the second study, then we focused more on the scope 2 emissions and yeah, we built simulation framework to compare drone-based operations and forklift-based scanning, you know, across different warehouses, size regions, you know, lighting assumptions and electricity grade factors. So, I would say that to summarize, what we did was, like, of course use, in this case, the protocol. You know, that's something that is standardized, validated across industry and also for research. First took a look at something more broadly, you know, about what was going on to better understand, you know, the conditions under which these drones work. And then, once we identified something more like an opportunity, then we deep dived into the scope 2 emissions, you know, and in particular for, you know, these inventory scanning activities.

- [Mackenzie] Today's episode is brought to you by the MIT Supply Chain Management SCM Master's program. Through capstone projects, companies can work with experienced MIT SCM students and faculty to tackle high impact supply chain challenges. In fewer than 10 months, your team gains practical solutions, strategic insights, and direct access to top talent, all through a project tailored to your business needs. Each student brings multiple years of real world supply chain experience with backgrounds spanning procurement, logistics, operations and more. Learn more at scm.mit.edu or reach out at scm-capstone@mit.edu. Can you all talk about one methodological challenge you face, taking us behind the scenes in the research process?

- [Camilo] Sure. For any model, it's crucial that you give it good inputs, right? So just to, you know, briefly again, describe the kind of inputs for these simulation, we consider different warehouse characteristics. So imagine the small, medium, large warehouses, you know, the amount of, the percentage of the area that was, like, allocated for pallet storage, number of shifts as well, hours, and well for drone inputs, you know, the number of drones, you know, depending on the type of facility and the different activities. We were very granular. I mean, we were capturing, you know, the speed of the drone, the time that it took to charge. Also when they will, you know, whether it's a power scan or more like in a turned area, the flight duration among others. So, I would say that for this and any other simulation or modeling, that they are crucial. And with our partners, in this case with Verity, they did this job very, very well, you know, in providing the information timely and precise such that we could, like, of course believe in the results, trust in them.

- [Mackenzie] In taking it now from looking at the simulated results and what the project revealed to real life implementation, in the capstone, you modeled scenarios where drone coverage increased from 64% to 90% to 100%. Can you talk about, in practice, what are the actual barriers to reaching that 90% goal? What's holding warehouses back? Why are we not looking at 100% drone implementation right across?

- [Tommaso] I think there are different aspects that speaks to this and different adoption barriers. So first of all, if you think of a warehouse of third party logistic players, you might have different clients, or other clients at a warehouse. And so maybe they have different operations, so they have a different willingness to automate. And so, this must be considered, that the solution maybe is not applied to entire warehouse. Another topic is also the technological limitation. So, the drones can scan what they see. So, our ideal use case scenario is a full pallet location, so you have one pallet per location. Because if you happen to have deep pallets, of course, the drones cannot see through. So I think it's a combination of technological limitations, business process change, cultural change, and also what are the kinds that actually that warehouse and what their desires are.

- [Mackenzie] You found that about 19% of warehouse locations can never be reached by drones due to layout constraints. How does that shape implementation strategy as well?

- [Camilo] Yeah, so layout and hardware, they really set the ceiling of what you can do in the warehouse. And of course, you know, there are technologies we always work to develop and make a technology better. And so, there is a way in which we try to extend this capability to different type of warehouses. Because, you know, sometimes we use the word warehouse, but when you enter into a single warehouse and then you go to another one, you really realize how important it is to have a site-specific calibration in this respect.

- [Mackenzie] The model that you all use shows that drone implementation actually reduces the number of personnel needed for inventory counting, but the overall workforce doesn't necessarily shrink, but it shifts. We're seeing that a lot with the impacts that AI is having across the board for different roles. What kinds of roles are emerging and how are warehouses managing the transition?

- [Tommaso] For sure, like, with AI, you know, this is a physical AI solution in the end, so it goes in the same direction. And one number that I want to mention is that in the US last year, the turnover of warehouse workers was 43%. And so, automation solution for companies working in warehousing, it's a way to address labor shortage rather than reducing or laying off personnel. And also another point to consider is that I've been around quite a few warehouses, and I've never met a person who was passionate about checking inventory. You know, checking inventory, it's a dull task, it's a repetitive task, it's a dangerous task because sometimes you do that at height, and ideally, nobody should be doing that. And so, what the technology really helps you to do is that you can shift the work of people that do a task that really is at high risk of automation and they learn digital skills, for instance, on how you ask a system of drones to do the inventory for you. And now you check the output of systems and you free up hours to do other tasks or checking other parts of the warehouse that the drones cannot reach, or doing different tasks on the controlled side. And so this is really the type of change that we try to build together with our customers.

- [Mackenzie] In the capstone, you ran several what if scenarios, including one where inventory accuracy did not improve with drones. Is there a point where drone implementation stops making environmental sense?

- [Elisa] There could be. All of these were hypothetical scenarios, as you said. What if. So, after we got our final 49.5 results, we were challenged to test the assumption that automation always helps. So, we were asking what if companies deploy drones but expected improvements never materialize? What would happen then? So, you reach a point where you have the equipment doing all the operations for you, you already have the energy savings and all of the other savings you could have, but if your inventory accuracy doesn't improve and if your material losses the other way, if they start increasing, there comes to a point where that loss and waste that we we were talking before offsets every other benefit. So, that is a hypothetical scenario, but it could happen and it's only to highlight the importance of, as we also mentioned before, the quality of the information. So, a system is only as good as the outcome it produces. So, you can have all the technology, all the process, but if it's not making your operations better, more efficient and improving the actual outcome, then you may not have all of these sustainability gains that you could expect

- [Mackenzie] Practically, looking ahead, based on the project, if a warehouse operator came to you and asked, "Should we invest in a drone-based inventory automation," what questions might you ask them before giving your answer?

- [Tommaso] I think it first of all depends on how much of their footprint is drone reachable. So the fit for the technology that was mentioning before. You also really try to understand in their operations where the pain points are. So, is it about the shrinkage, the write-off, which is probably a bigger problem from manufacturers that it is for third party logistics, because they produce the goods that that they lose. So, they even bear higher economic consequences. Or if it's with the partial deliveries or what they need to deliver the product, they cannot find it. And then, I think it's very important to ask them if they are measuring this data, so which KPIs they want to change. And of course, you know, drone solution is something that works very well where you have a warehouse with very high ceilings. And so, again, the idea that not all warehouses are the same and so it's really very important to look at all these elements together.

- [Mackenzie] And you recommend treating drone coverage as a master lever for emissions optimization. Can you talk about what that means operationally and how leaders should think about phasing implementation?

- [Elisa] So usually, and more nowadays, companies and leadership might look at automation, AI and think the more, the better. So, I can say if I am fully automated, that means that I'm gonna get all the possible benefits, and that is not always true. So, what we were trying to say is that more coverage isn't just about how much you have automated, but also that it influences multiple process simultaneously. So, we saw that it's kind of like a chain reaction, right? Drones, they have better issue rate or the issue rate got better when drones were counting the inventory rather than humans because they're more precise, they do the counts more frequently than humans could possibly do. So they are better at catching errors, right? So, because they were able to cover more locations, then the accuracy got better, the issue rate decreased, there were fewer issues, fewer write-offs, therefore lower emissions, right? So, from an implementation perspective, this means that leaders shouldn't view coverage as a simple technology metric, but more they should think about it as a strategic variable that is gonna shape the overall performance of a warehouse. So, the more the drone covers, that means the better results you're gonna get.

- [Mackenzie] Beyond drones, what other practical recommendations would you make to companies looking to reduce scope 3 emissions?

- [Camilo] I would say that the most important part of this kind of project is to allow managers and people to look at other types of technologies, to be open to try other things, right? That could, in this case, cut carbon emissions. So essentially, when you talk to some practitioners, they have certain ideas on how to do things, how to mitigate, you know, carbon emissions, how to follow what others are doing across your sectors or industries. But in this case, I would say that one of the most important parts and one of the most, the things that I like the most about this project is innovation, right? So, imagine, this is very futuristic, right? Like, now, instead of forklifts, using drones to scan inventory and guess what, after doing this and the simulation, these yields, you know, these significant reductions, that's amazing. Maybe this is an opportunity to, you know, think outside the box and try other things and instead of, you know, an AB testing, well, I invite, if you are listening to this and you want to try this, well, reach out to us and maybe we can conduct research together to really understand whether these solutions are good or not. So that's, I would say, you know, in terms of solutions, the way that I would recommend people to try in their industries.

- [Mackenzie] And Tommaso, from Verity's perspective, how are customers actually using the sustainability data? Is it driving purchasing decisions, or is it considered a nice to have?

- [Tommaso] So the the honest answer to that is that sustainability is increasingly part of the conversation, but it's not yet the sole driver. And of course, there are some nuances, there are some differences between Europe and the US. And within Europe, we see that this is particularly important for some kinds in the Nordics. For instance, one of the climate partner for the second study on scope 2 emissions, they run a sort of internal carbon pricing where they charge their own countries based on their emissions. And so, for them, when they see a reduction in CO2 emission enabled by automation solution, it's much easier to attach an economic value to that than it might be for other clients. And so, in some cases it's easier, in some cases it's less easy. But there are two other things I want to mention and I think we are really supporting company in the right direction. And the MIT report from last year on state of supply chain sustainability, they noticed that 70% of firms, they say the lack of supplier data and especially scope 3 reporting, is a major limitation. And most of them still rely on a spreadsheet or some accounting tools. And so the additional benefit solution like ours on top of the reductions, that you have that evidence and visibility layer, once again, where you get all the role and timestamped activity data on your inventory, what you lost and so on. And it is also sort of a data equalizer opportunity because even a smaller company can then report and trace their operations and their stock at the same level of the global corporate. And so this is also the contribution that we're trying to make.

- [Mackenzie] Elisa, you conducted interviews with warehouse teams to validate assumptions during the capstone project. For you, what was the most important insight that you gained from talking to people actually doing the work that the data alone may not have told you?

- [Elisa] Yeah, that was a very big part, because as Camilo said before, one of the biggest challenges is representing reality correctly. Like, we can make a thousand assumptions and then you go and talk to the people who's actually there using the technology and doing the work and they paint a completely different picture. So, it was very interesting learning how the actual operation worked, how they had to go and find every single missing item, every time they find a discrepancy, they had, like, this whole process of, like, a team of people going location by location, tracking the system, where was it last, where could it be next? And then it's a whole chain reaction of manual efforts that require equipment, people, energy and time. It's very time consuming. So that was very, not surprising, but that really helped us shape how we were gonna do this methodology and how we were gonna represent that effort into numbers and eventually into emissions, right? Another important thing was that it was very good to know how the operations and processes, they all depend on the warehouse, their consumer, their materials, their regulations. Like, we interviewed two different warehouses in two different locations and they operated in a different way. Not a complete different way, but they had different customers and that brought different requirements. Some of them had contracts that required the materials to be in a specific aisle that could not be accessed by drones. The other one didn't have that. Also, some were storing their items in cases, other in pallets, other in units. So you need to know all the specifics to be able to try to quantify this kind of thing. So talking to the people, that's always very, very important.

- [Mackenzie] Looking to the future of the work. So, Tommaso, this capstone was published last year and Verity decided to expand and do another capstone this year. Why did you all decide to do a second project and what did you change this time, or what more are you looking to find?

- [Tommaso] In this first project, we were looking at emission across all three scopes, so it was a bit larger in terms of the type of emissions that were considered. Whereas the second project is focused mostly around scope 2 emissions. But at the same time, the first project was really looking at a single warehouse with a snapshot of a real case study. Whereas in the second project, we moved from a single warehouse to a portfolio warehouse. And what we did was more a simulation model that could drive decision making for implementation. And so I think, you know, the separate answer, two different questions, in one case, can we quantify in a real case where we implemented the drones, what were the benefits? And second was more, okay, we want to implement the drone in portfolio warehouses, which one should we prioritize? Which countries need this first? Where can we maximize the impact on the portfolio of clients by prioritizing certain types of warehouses or certain countries with different carbon emissions? So, we were answering different questions and different needs of the clients that were involved in this project.

- [Mackenzie] Could you all speak to the benefits for all the stakeholders involved in a capstone project and perhaps distinguish how it's distinct from, say, a consultant firm?

- [Camilo] So I think that there are various stakeholders that benefit from these kind of projects. Of course, students, you know, because they gain hands-on experience by working on, you know, these real world problems with industry stakeholders. And as a consequence, they develop technical, analytical, communication, project management skills, right? While, you know, also helping, like, the students to translate the analysis into actionable recommendations. In addition, of course, companies, sponsored companies, because they obtain rigorous analysis on their problems, right? And this usually results in developing practical tools, models, dashboards, frameworks, so on, so forth. And certainly, for us as researchers, well, it's an opportunity to bridge the gap between theory and practice while, you know, while producing insights that become publishable, you know, with papers and of course practically relevant that we can disseminate to, you know, improve the conditions of society. We are doing this because we want a better world, right? So, in terms of sustainability, well, this is the case of this company that is interested in doing this and that, and by doing rigorous research with the most talented students in the world, now we can show, you know, the results create impact beyond, you know, a single company.

- [Mackenzie] As Verity is doing a second capstone project this year, could you all speak to some of the people involved in working on the project?

- [Camilo] For sure. I would say that the work, you know, that our two talent students conducted was amazing. You know, I'm referring to Liem Phan and Shrivats Agarwal, they did an amazing job, extremely committed to the project and, yeah, again, kudos to this work. I would say that in this case, the sponsors, you know, were very happy with the deliverables and they built that, I don't know, a communication, a strategy and environment such that they were, you know, our sponsor is willing also to share many things that, as Elisa also mentioned before, you know, it's important to model things that are real and that are practical, that they are useful, right? So, from the very beginning of the project, they were, I would say, committed to that. And yeah, so to them, and of course to Dr. Josue Velasquez, who I would say was the mastermind, you know, behind all of these two projects, you know, connecting also with Tommaso, you know, positioning the different studies and again, you know, to create impact. So yeah.

- [Tommaso] I would say this speaks also to the benefits, right? Because since the sponsor of the capstone, what we bring in is our time and our commitment, but also we bring in operational data from actual warehouses that we make available, that our clients make available, that would be otherwise difficult to access. But we also gain access to incredible talent. We were extremely lucky last year with Elisa and Philip for the case study on the single warehouse, and this year with Liem and Shrivats for the simulation that we did in collaboration with DSV. And then, of course, you know, it is very different from a commission consultancy side because of course there is a greater risk here. I mean, we were very lucky that we very much liked the results of both the capstones, but that's not granted because the additional measure that you get is that you get a stamp on these results from an academic institution that doesn't compromise on anything. And so, it is a guarantee for us, a guarantee for our clients that the numbers that we're getting, of course, always with assumption, with all the limits that balance with any that are acknowledged, but really indicates a strong direction that you cannot get from any other source, I would say.

- [Elisa] I can add something from the student perspective. I think it is very valuable because of all of the reasons Camilo and Tommaso mentioned, but also there is a big partnership between the researchers, the sponsor company, and the students working in this together. Like, you can really feel how everyone is really invested. They all are expecting and working towards the same goal and really the best outcome possible. In this case, the project was mostly aimed towards sustainability, but for example, because we were also working with automation solutions, we partnered with Dr. Miguel, who specializes in automation, and I think that helped us kind of like have a team that is really an expert in sustainability efforts, but also in automation. And you have the company that brings all the technology knowledge, solution knowledge, and you have the students that are gonna help put their own perspective with all the things they're learning, like modeling methods and AI coding. So it's really a great team for all the capstones, not only in my experience, but for all I've seen with the rest of the cohort. Everyone had a great experience working with the researchers and their sponsored companies.

- [Mackenzie] That wraps up this episode of "Supply Chain Frontiers". A big thank you to Tommaso Portaluri, Camilo Mora and Elisa Ruiz for joining us. I want to credit Philip Kook for his work with Elisa Ruiz on the 2025 Capstone project as well, which you can read in full under our publications at ctl.mit.edu. "Supply Chain Frontiers" is recorded on the MIT campus in Cambridge, Massachusetts. Our sound editors are Dave Lishansky and Daniel Simpson at David Benjamin Sound. And our audio engineer is Kurt Schneider of MIT Audio Visual Services. Our producer is myself, Mackenzie Berry. Be sure to check out previous episodes of "Supply Chain Frontiers" at ctl.mit.edu/podcast or search for us on your preferred podcast platform. I'm Mackenzie Berry, thanks for listening and we'll catch you next time on "Supply Chain Frontiers."