Truck and Drone Cooperative Delivery Vehicle Systems - Social Implications

Arthur:
Welcome to MIT Supply Chain Frontiers, where we discover the future of global supply chain education, research and innovation. Brought to you by the MIT Center for Transportation and Logistics. Every episode features center researchers and staff who welcome experts from the field for in-depth conversations about business education and beyond. Today, we're speaking with Fari Gabba, research associate with MIT CTLs, mega city logistics lab.

Fari Gabba:
My name is Fari and I am currently a technology and policy program student at MIT. I work in the mega city logistics lab under Mathias [Winkenbach 00:00:42] and we do a lot of work in last mile logistics, optimization, operations, routing and planning. Recently Mathias and I have been working a lot on drone delivery operations and what that means for last mile logistics going forward.

Arthur:
Why do you think this is an important thing to know about?

Fari Gabba:
Last Mile represents a disproportionately large share of the overall delivery supply chain cost, and so drone delivery seems to be a solution for logistics. We tackle drone delivery in two different ways. One is the airplay drone operational model where we have drones operating on their own. The other model is what we call a truck and drone cooperative model. And this model is really about drones that sit on top of standard delivery trucks and they sort of go out into the city and do their work in tandem and try and extract some efficiencies from doing work in parallel rather than a single driver doing single deliveries at a time. And so this second model is much more aircraft carrieresque in its approach to Last Mile delivery. And we think that these two models are what the future of Last Mile delivery look like. With all of that said Mathias and I are sort of realizing the value of understanding the broader social and economic and regulatory implications of drone delivery beyond what we typically work on, which is routing the drones to the customer between the depots.
There are many technologies that we've seen that have almost been deployed prematurely. Societies are actually having to bear some of the negative externalities of these technologies. The most common example of this technology policy conundrum is social media and its capacity to propel divisive news. And so there are many examples like this, where we need to really think about the broader implications of new technologies. In the logistics space, firms are working on a variety of different technologies, even beyond drones, just sort of meet the Last Mile problem head on. And we think that there's much value in understanding that the whole system and painting a full picture of what this technology would mean for our cities, for our lives and for the citizens they're trying to serve.

Arthur:
You recently published a working paper with Mathias and the Megacity Logistics Lab. What does mean to have a working paper?

Fari Gabba:
A working paper sort of means something that we envision will be part of our feature research going forward and something that we're still sort of ruminating over and trying to develop the concepts around. This paper sort of tries to address what this technology would mean for cities and societies, if deployed at scale, and what should the stakeholders be? Be it the logistics firms, the regulatory bodies, citizens, municipalities, what should they be thinking when they talk about, or when they hear truck and drone delivery? In the end, when a logistics firm or some people working in this space have to route a drone from their delivery point to and from their depot, they won't be able to fly in a straight line. They'll have to probably avoid certain areas or have these sorts of considerations in mind, and that will have implications for their bottom line and their operations. And of course, for the impact they have on the communities they're serving,

Arthur:
What are the biggest challenges that you're facing in researching a topic like this?

Fari Gabba:
Yeah. So that's an interesting question. Well, let's break it down to two things. I guess the first one is let's understand what the challenges are for the technology specifically. And then we can talk about what are the challenges for researching such a technology. So the former, what are the challenges for the emerging technology, the truck and drone cooperative system, mainly regulatory in scope, and actually social in scope. So we found that the barriers to the truck and drone cooperative model are actually quite similar to that for typical, pure play drone delivery in that the regulations are not really there to support these sort of operations at scale. Operational safety integration into sort of the larger airspace, some of the privacy implications of drones flying overhead and having active cameras and actually some of the noise pollution and noise impact that these sorts of operations will have will be barriers to operations in specific regions and municipalities, but also at a national scale. For the truck and drone cooperative model, specifically, drones will have to land and take off in and amongst residential areas and public spaces.
And so maybe the noise impact or many of the other negative externalities of such a technology will have a larger bearing on the people around them because of how integrated it is into the city. So, that's sort of the first dimension. And then the second dimension is what are the challenges we're facing in researching this technology? So these technologies, I think, are very well studied in literature from a routing perspective and from an optimization perspective, but having to integrate the two dimensions of our analysis, the qualitative socioeconomic regulatory perspective with the quantitative dimension is actually quite hard. And that's something that we are thinking of doing as we go forward, because this is what we really care about is how does this technology fit into the world? And, what will the system like in the future, in its entirety, not just in a closed solution.
And so integrating these two dimensions is not easy and it poses a large challenge for us, because it's hard to quantitatively model some of the softer impacts that we're trying to understand. For instance, if we think about privacy, how should we route a drone to minimize the privacy risk or the risk of privacy breaches, and how do we quantify that risk? And so these are the sorts of challenges I think we're going to be facing going forward. And I think this is not new to research. I think it's unique to all the stakeholders involved. These technologies are so disciplinary and complex that these questions will need to be asked and we'll need to have answers.

Arthur:
What are you digging into right now with this piece?

Fari Gabba:
What we're currently digging into or currently trying to decipher is what are the key social regulatory barriers we can really get our hands into and which ones are worth modeling on a quantitative level? This sort of analysis will hopefully inform some of the quantitative models we begin to develop as we move forward. Aside from that this working paper also attempts to devise a suggested regulatory framework for these sorts of technologies. And we try to use planned adaptive regulatory framework, which is common to most emerging technologies that we see today. And so this working paper sort of has two dimensions to it. One is the implications of the technology. And the second is what can we do to mitigate these implications from a regulatory perspective? And hopefully these two analyses will help us inform some of the modeling that we do later on.

Arthur:
That's excellent. Great. And so then what keeps you up at night? When you're working on a big longterm project like this, I've find that something bugs you, something eats at you. Is there anything here in this that is bothering you?

Fari Gabba:
I think something that keeps me up at night with drone delivery are just the variety of implications that it does have for our cities. And I think the different implications it will have for different cities and different people around the world. No city is the same. They all have their different challenges and unique landscape, and of course, culture and people that are actually living there. And so whilst we can, from an academic perspective, we can say these are what we think the major concerns will be going forward, I think it will be hard to really comprehensively grasp all the different dimensions of such a technology as it gets deployed at different places around the world. What keeps me up at night is if and when we decide to sort of integrate this into a quantitative model, I hope we don't miss anything, is really what I'm hoping for. And that we are able to build a system and a model that is as representative of what we expect to see in the future as possible.

Arthur:
Where do you see this headed?

Fari Gabba:
This research has really opened our eyes to the ability of our lab. It's unique in the sense that we have a long history of quantitative modeling, but we have this passion for the broader implications. We see this sort of research being more integrated into our future, where people might find it interesting that we really think and believe that drone delivery is very feasible cost-effective, economically, socially, environmentally sustainable in less densely populated areas. And so whether that be suburban or rural, and we think that from both dimensions, it all makes sense because on the social dimension, we see a lot of those barriers I mentioned earlier are being alleviated and on the quantitative modeling and economics side of things, we see cost benefits being extracted there.
So at least in the short term, I envision this sort of research informing some of the decisions that logistics firms and partners in that space are willing to make about where should they deploy their drone technology. And I think in the longterm, I see this sort of research being applied and applicable beyond Last Mile logistics. I think it's pivotal for research in supply chain more broadly to really understand the technology policy and the social implications of the technology that we're trying to implement.

Arthur:
And so we have a lot of people associated with the MIT Center for Transportation Logistics who are in industry. They don't really get a chance to take a deep dive the way you're doing. Do you have any recommendations or ideas that you can give people who are trying to get some answers outside of academics?

Fari Gabba:
Yeah. I think that if I were to recommend something to try and understand or ask some questions about these sorts of Last Mile logistics or broader supply chain questions, I think what was very valuable to us was being able to reach out to people and incorporate some of their ideas. And this is sort of beyond MIT. I think we reached out to a number of our logistics partners, industry partners, and to try and understand how they see it and what their perspectives are. We gained a lot from working with the political science department at MIT, and I think that it was valuable for us to keep an open mind about who we approached, about what type of ideas we have simulated, because oftentimes I found the most valuable insights were from those who are almost not even tangentially related to our work.
That was sort of a key finding I found. And it's something I'll work on going forward is keeping an open mind with the questions I ask, the approach that I take, and the outcome I'm hoping to arrive at. And I guess the ideas and experiences I try and assimilate with my own, for industry partners and people in industry hoping to explore this sort of problem further. I would suggest reach out to CTL of course, but also be open to assimilating different learnings and different ideas from different fields, because although not unique to last mile logistics, these technologies are incredibly interdisciplinary nowadays, and it's not unique to one field of study as we know it. And so being able to assimilate different concepts from different fields is going to be a very valuable skill to have going forward, especially in answering these sorts of complex questions.

Arthur:
What's your approach going to be? You've got all this quantitative stuff you want to do, and you have all this soft, personal behavior you want to measure. What orientation will you take next?

Fari Gabba:
I think we need to explore the field a little bit more because in Last Mile Logistics, we haven't really seen this being modeled per se. I mean, the last mile literature is deep in the optimization and operations research sort of fields, but we don't have too much of the social and broader implications. So I think being able to assimilate some of the learnings from different fields, we can attempt to try and model these and we see what the implications or what the results are. And we try and understand to do this, makes sense to us on an intuitive level. And let's try them out in different use cases. And the more we train our model and the more we develop our thinking, I think the closer we'll get to being able to model these qualitative factors. But as I said earlier, there's no right answer. And so that's what's, I think, very different to all the work we've been doing in the past is until now there's always been a number or right answer that we land on.

Arthur:
Absolutely. I like questions with no right answers. They're real-life questions. For a working paper? Are you looking for feedback? Are you looking for people to say yeah, but what about?

Fari Gabba:
Yeah, absolutely. That's a very good point. We want to know feedback. We want to hear feedback from people in the Last Mile Logistics space, but also beyond it, because I think to be frank, that was one of the most, I guess, pivotal dimensions of this paper was our ability to reach out to other people at MIT and other researchers and industry partners that we have that are passionate about drone delivery, but not necessarily involved in that space. And so feedback about some of the implications that we could be missing and potentially some of the solutions that we propose, I think would be valuable for our research going forward. And also just for others who are exploring this and others who are hoping to implement this going forward. This is part of the exercise of writing a working paper. And we very much look forward to hearing feedback.

Arthur:
I want to thank you a bunch. Thank you for taking the time to talk about the paper. And it's fascinating, I think, where you're headed.

Fari Gabba:
Thank you very much, Arthur. Speak soon.

Arthur:
All right, everyone. Thank you for listening. I hope you enjoyed this edition of MIT Supply Chain Frontiers. My name is Arthur [Grall 00:14:54], communications officer for the center. I invite you to visit any time at ctl.mit.edu or search for MIT Supply Chain Frontiers on your favorite listening platform. Until next time.