Eva Dale 0:00 From the heart of The Ohio State University on the Oval, this is Voices of Excellence from the College of Arts and Sciences with your host, David Staley. Voices focuses on the innovative work being done by faculty and staff in the College of Arts and Sciences at The Ohio State University. From departments as wide ranging as art, astronomy, chemistry and biochemistry, physics, emergent materials, mathematics and languages, among many others, the college always has something great happening. Join us to find out what's new now, David Staley 0:32 I am joined today by Morton O'Kelly, who is Professor of Geography, and the Divisional Dean for the Social and Behavioral Sciences at The Ohio State University, College of the Arts and Sciences. From 2012 to 2015, he served as Director of the Center for Urban and Regional analysis, and from 2003 to 2011 as Chair of the Department of Geography. His areas of research interest include network analysis, logistics, transportation, and spatial interaction models. Welcome to Voices, Dr. O'Kelly. Morton O'Kelly 1:04 Thanks, David. David Staley 1:05 You work in the area of transport analysis. And I'd like to begin first by giving us a definition of transport analysis. Morton O'Kelly 1:13 Well, we look at transportation as an important part of the economy, it helps move people, goods, and it is a vital way to connect what we think of as separation in the economy between sources of materials, and sinks, or places where those materials are needed. And unfortunately, typically, those are not in the same place, so we need transportation to overcome the separation. So it's a very geographical issue, it's a very spatial issue, as we call it, and it lends itself to thinking in an organized way. So instead of having sort of random flows, we have very organized flows. So transport analysis is all about trying to uncover, say empirical regularities or normal aspects of transportation, and then ideally, once we understand and describe transportation, we'd also like to try to optimize or make it as good as possible. I have a little phrase I teach in my class, you know, nobody wants to waste money on transport, it's not a good thing. But you have to spend at least a certain amount to overcome the separation. And so efficiency means, let's drive that down to the bare minimum. David Staley 2:20 And we're talking about the movement of things. But also if people. Morton O'Kelly 2:23 Yes, certainly, both people and freight movements are very important, and in packages and specialized transportation of, say, supplies, or emergency supplies during periods of need, such as an outcome of a hurricane. So I like to think of a typical example, let's say you have a region where, by good fortune, you're able to grow lots of cherries, and cherries are a great product, and they're made in huge volume at a certain time of the year. But you can't potentially sell those to the local population, you need to have a much wider market. So moving those supplies from a source region, say Michigan, to the rest of the country becomes a transportation problem. David Staley 3:03 Well, I know one of the problems that you work on is commuting, movement of people, I suppose, I know you did a paper recently that asked if cities with longer commutes are necessarily inefficient. And I saw that and I thought that must be controversial, because I assumed that we don't like cities to sprawl. Morton O'Kelly 3:21 Exactly. And we thought of that topic and the title in a way to attract attention, because normally you'd think David Staley 3:28 What was the title? Morton O'Kelly 3:29 Well, "Are longer commutes necessarily inefficient?". And so, when you have a large city, I think Atlanta is the poster child for this, a very large city where the commuting times could be in the 30-40 minute range. And you think, wow, that's very inefficient. Well, in a sense, it's not great that they have such long commutes, it's an inevitable consequence of the metropolitan form where counties, outlying counties are sending people through the interstates and through the roadways to their places of work, but it's not necessarily any efficient, because the people are making almost ideal shortest paths to their destinations. And so, I like to use a little thought experiment, imagine a long river valley, and there's a steel mill at one end, and there's a residential district at the other. Those folks are going to have to travel up and down to that mill, there's no other way about it. So in that sense, the efficiency for them would be they're spending just exactly what's needed to get to work. Atlanta, I won't say is ideal, but it is close to, given the current layout of population and workplaces. It's closer to efficiency than, surprisingly, a city like Las Vegas, which has a much shorter commuting range. It's a much more compact city, but people there are more or less higgledy piggledy, traveling all across the cityscape and making comparatively less efficient use of their journeys. So I'm not claiming Atlanta is ideal by any means, it'd be much better if we reorganize things to be more concentrated, but it is a little bit of a surprise when you take a Vegas, Atlanta contrast and see that in one respect, Atlanta comes out, at least as trying its best given the circumstances and Vegas is not doing so well. And you might ask then, Well, where does Columbus lie in all of this. And, David Staley 5:16 Indeed. Morton O'Kelly 5:17 We did a little chart and we took 25 or so cities and categorized them into which of the various bins they would fall. And as you may realize, from Columbus, we're somewhat blessed by having a relatively moderate commuting distances, and relatively efficient organization spatially. So it was a little bit of a technical paper, the mathematics of how to get that efficiency measure a bit tricky, but the overall idea was to try to categorize and scale cities along this dimension of efficiency and commuting range. David Staley 5:47 Maybe you've gestured to this already, but efficiency means what just speed of the commute or...? Morton O'Kelly 5:53 Well, speed and movement, people are happier when they're moving at freeway speeds, of course, but efficiency would be if I gave you a budget to make a connection, would you end up needing to spend more than that, or would you be able to survive with just the minimum? So, an ideal efficiency situation would be where people are able to make trips that connect to close by places of work. So historically, many cities did have almost walking range to factories and to places of work. Nowadays, we have suburban sprawl, we have even ex-urban sprawl, and potentially people making some quite extreme commutes. We're not celebrating that, we're just saying that, you know, this is a current situation based on the way the city is laid out. And then that leads to great questions about sustainability, should we think about reorganizing cities to have workplace and residential place, in a sense more closely choreographed. And some trends in that direction are happening where there's infill development in cities, people are moving back to be closer to work, perhaps even being able to cycle or take the bus to work. And we're all about that, we're interested in efficiency and then realizing, wow, some of these extreme commutes are, in fact, quite concerning in terms of the amount of resources that are required. David Staley 7:13 So you have some other cities here, that you looked at in this study. Morton O'Kelly 7:17 And I think it's useful to think about and probably the, the popular imagination already considers this, that a city like San Diego turns out to be both inefficient and sprawled. And this is partially due to geography, it's a coastal city, there are places of work, large naval yards, airports, universities, and quite a bit of spread. So, if we were to do a classification of cities, in some ways, San Diego would be worse than us both in terms of journey length, and in terms of efficiency. And then just to fix ideas, are there any places more efficient than us? Well, Wichita, Kansas actually has both compact city and efficient use of where people are using those resources. So there are two examples from the extremes. David Staley 8:03 Did your research at all consider the type of transportation, car versus bus, or subway or...? Morton O'Kelly 8:10 Absolutely, yeah, so we would be interested in moving more of our commuting public from car dependence to sharing or to public transit or to other uses, other modes of transportation. We've had limited success with that in some of the cities we've studied, but in the northeast, there are certainly large, compact, dense cities, which favor the use of transit. So there's sufficient density to have a very high subway, ferry, bus transit options available to us. Columbus, again, is kind of in the middle, as I've mentioned. David Staley 8:46 What would it mean to make a city more efficient, what would be involved in that process? Morton O'Kelly 8:52 So planners have certainly looked at this idea, and have come up with ideas around co-locating or closer location between your residence and your work. So if, of course we work at OSU, we have to get here from someplace. What if we were able to live within closer range of the university, or if there was a closer connection between the development of suburbs and the provision of workplaces, shops, and services in that region to make it more compact. And some of my colleagues in geography have really preached a lot about thinking around changing the organization of cities to being more efficient and more friendly towards transit use, which I favor of course. David Staley 9:34 So you have another area of interest around the geography, transport analysis of hub and spoke air carriers versus point to point carriers, and first of all, point out the difference and tell us a bit more about the research. Morton O'Kelly 9:48 Well, thanks. That's been one of my Hallmark research areas for 30 years and, the idea of point to point is best exemplified for everybody's understanding by Southwest Airlines. David Staley 9:59 Southwest, Morton O'Kelly 10:00 They typically will fly you from here to Baltimore or from here to St. Louis, on a flight that's probably doing just that, going back and forth between pairs of cities on a relatively short range over the course of a day. And, by the way, we, David Staley 10:15 So no connecting flight? Morton O'Kelly 10:16 Correct, you can connect with them, but that's not how they're set up to organize themselves. So perhaps you would go through one of their connecting nodes on the way to your destination, Midway being one of their cases. I'll come back to that in a second. In contrast, we have airlines like Delta who operate a hub and spoke system, and they make their living by mostly channeling you to a hub and having you connect to their connecting flights and make the rest of your journey on another Delta flight. David Staley 10:47 Atlanta or Minneapolis, Morton O'Kelly 10:49 Precisely. So Atlanta is a mega hub for Delta. And this contrast between point to point and hub and spoke networks has fascinated me for a long time, and I've only recently begun to really get to the the nugget of this, and Southwest again provides an excellent example. I just have to do a brief detour into a concept that we have called a triangle inequality. And this is, draw a little triangle, and if you go from A to B, you're making a short path from A to B, as opposed to going A to C to B through some intermediate point C, well, Southwest, bless them, they charge you less money, if you go directly from A to B, and they charge you a little bit more money if you go through C to get to B, as you'd kind of expect, logically, you know, they're going to be holding you further, they're carrying your bags, it's not a big surprise to think that the A, C, B trip should cost a little bit more. And indeed, if you go on Southwest Airlines sites, and I've done this recently, you'll find that to fly to Boston directly is actually less expensive than if they take you through Midway or some other city. All that's fair. And it's kind of how the public thinks air fares should work. In the case of Delta, however, it's the exact opposite, they are likely to charge you a premium for going on a non connecting flight. And they may charge you much less for going through an intermediate node. And you say, How could that be, how can they make money doing this? So the logic of air transport networks and hub and spoke is such that Delta is in the business of making a dense flow. They're trying to bring a large number of people to their hub. And they're collecting you to the hub, and they're going to offer you a convenient connection there to get to Tallahassee or Wichita or wherever you're going, and they will make a larger flow of people say from Port Columbus to the hub. And, understand that you'll have to change flights, but you'll get to where you're going, and the beauty of this from their point of view is that then they have a, what I would call a fatter flow, a larger volume that allows them to optimize the aircraft size, the times of departure, the frequency, everything that you could need to make the airline operate is set up in their favor. So in a way, by offering you a slightly lower fare, in fact, a good deal lower fare to go through Atlanta. They're offering you sort of a share in the savings that they get from operating a dense, efficient network. We've done lots and lots of studies on this. And one way that it comes down to bear is, well what aircraft is the airline actually using? Again, I'll use an analogy. Many of us have heard the term gas guzzler. So if you have an old car and it's a gas guzzler, well, it's going to cost you a little bit more to run it, is not a great big deal if you're not necessarily going to go out and sell that and get a more efficient vehicle. Airlines, on the other hand, are running aircraft and some of those aircraft are gas guzzlers. There's an aircraft called a Boeing 727, which is no longer in use, which had three engines on it. And it was a notorious, heavier user of gas then say more efficient aircraft. So, as the fleet is organized, the airlines would love to use their most efficient aircraft. But, quite simply, they don't have enough and they have a fleet that's made up of a mixture of many different aircraft types. And they will continue using MD-80 or older aircraft because that's what they have available. And one of our studies then said, Okay, let's look at which flight legs are particularly out of whack with what we would like to do. And so when you look at some aircraft results, we're able to find that, for example, let's say an American Airlines scheduled route between Dallas and Norfolk, Virginia, using an MD-83, would be much much better if they could use a 737. And they would love to do that, they need to get more to be able to do that. And this brings us to a current topic that's in the news and that is, how come Boeing got in trouble with their 737 Max, and it's all about their attempt to design more powerful, more efficient engines and put them on an aircraft that would give airlines a fuel consumption advantage. It's an overwhelming advantage, the airlines would love to run planes that are not gas guzzlers, but in doing that, you need to be careful to make that plane safe and controllable in the ways that we've seen talked about in the news, and that's an ongoing topic. Eva Dale 15:24 Did you know that 23 programs in the Ohio State University College of Arts and Sciences are nationally ranked as top 25 programs with more than 10 of them in the top 10. That's why we say the College of Arts and Sciences is the intellectual and academic core of the Ohio State University. Learn more about the College of Arts and Sciences that osu.edu. David Staley 15:49 Has this research influenced the airlines? Do you work with or consult with their lines in this research? Morton O'Kelly 15:55 Great question. And honestly, I'd have to say that, despite having spent this amount of time on my work, I have not consulted with the airlines on this. In fact, this is more like an academic, almost purist, think of an abstract way that we could study this problem and do not do it as a consultant to American Airlines. Now, I will have to say that the airlines are quite capable of studying these problems themselves. So some of the biggest employers of people who work in my area of operations research would be the airlines, they have a very active effort to optimize everything to do with demand, with fuel consumption, with fleet usage, with the decision about well, what size aircraft should we fly between here and our destination? What should be the ideal? And we do that more from a abstract modeling point of view to set a publication that will say, well, here's how we think it should go. And I would bet that the experts in the field within the industry would be able to say, Well, that's all fine, but we need to add these other considerations. So they have additional complexities even beyond what we can study. One would be the capacity of gates at airports, you simply cannot move large aircraft in and out of some airports without great difficulty. And so we have an almost an idealized, pure version of what should happen, then it's up to the real world industries to figure out the optimal way to do it from their point of view. David Staley 17:18 Any applications beyond airlines? Morton O'Kelly 17:21 Well, I'll mention freight actually, I have done a lot of work with FedEx in terms of their data. So FedEx, just as airlines are moving people, FedEx are moving packages, and they have a wonderfully efficient and choreographed system that operates to connect a global network of cities. And many of the same questions come up in the case of FedEx, what aircraft to use, what are the most efficient ones to use, but one great advantage they have is that they fly in and out of Memphis, Tennessee, and they primarily have their peak load in the middle of the night. The aircraft come in, they unload the aircraft, swap them all around, and ship them back out all in the period of say, midnight to 4am. And so, to their advantage, they are the only operator using that airport at the time, so they can be ultra efficient in making the best use of technology, making the best use of their fleet, making ideal uses of sorting and techniques to make a minimally costly operation. So as I mentioned in the beginning, no one wants to waste money on transport, including the carriers, and so, the more that they can be efficient and well organized in their system. That's why I love that as a project, to try to understand well, what exactly is going on here? So we found some fascinating things where some of their routes are sufficiently busy that they would be flying two or three aircraft per day over that segment, and interestingly, not all the same aircraft. So this would be an effort to try to make the best combination of structure and size to give them an ideal outcome at the lowest cost. David Staley 19:01 I know some of your work also involves historical aspects of transport or historical developments. And as an historian I'm particularly interested in this, so tell us about this research. Morton O'Kelly 19:12 Well, this is actually more of a kind of a pet project and a background hobby of mine over the years where I've taken a great fascination, some might say, you know, too much of a fascination, with canals because canals, let's go back to the mid 19th century, they revolutionized the way materials were transported to cities and provided a big focus on a way to change the organization of flows. I use the word organization again, you know, how the flows are happening, was modified and has had a resultant change from that historical period. David Staley 19:49 Like human made rivers. Morton O'Kelly 19:51 Exactly, and it, in many cases, solved connection problems and provide a connection. I've studied the Illinois and Michigan Canal in detail, but I also found this great quotation to do with Ohio to connect the Miami and Erie Canal, Antigua, and this canal was written about by one of my, I would say forefathers, the original Chair of the Department of Geography, a guy called Clifford Huntington. Charles Clifford Huntington wrote, in 1905, that before this canal was opened, there was virtually a local economy, you would grow corn or whatever, and you'd have to consume it for yourself, there was no market. And this was true up to the period of say, roughly 1848, and then it said, by the end of 1848, the wilderness district was in direct communication with the northern and southern markets. And then, three years before, not a single barrel of flour or pork, or a single bushel of grain found a market beyond the immediate neighborhood, but 1848, and this is in their dollars at that time, $400,000 worth of those articles were produced and transported by canal, mostly towards the north. That's just one simple example where we're talking about the area in western Ohio, where there was a convenient, relatively flat landscape for canal building. And, by way of connecting the Miami River to Lake Erie, provided the farmers, once they clear the land and provided copious quantities of inputs to the land, then found a market for their goods, they were able to sell those goods, the money from those sales came back and allowed town building and silos and farm expansion, and many, many developments in the property and land use picture of the state of Ohio. The exact same thing happened in the state of Illinois, when the relatively short canal joined the Illinois River, about 96 miles segment of Canal joined it to Lake Michigan, and that opened the entire central portion of Illinois, and its product, in this case, corn primarily, to be shipped to the lakes, where it's not a trivial matter then to get it to eastward to the rest of the country. But it was much much much better than it had been before. And so Illinois just blossomed in terms of the production of corn and other crops for export. And in turn, the canal provided a transport mechanism in the opposite direction for furniture and farm machinery and stone and heavy building materials that allowed towns to be put together. David Staley 22:23 So you study transportation, supply chains, logistics, as I hear this, I think of someone that maybe is in the Fisher College of Business, or maybe an economist. How does a geographer study these things? Morton O'Kelly 22:35 Well, that's a great question. I did my undergraduate degree in geography and economics, so I think it's correct to think that I bring some of the flavour of reasoning about the logic of things to this. It is true that many of my colleagues who study these types of problems are in business colleges and in management schools. And I think that's fine, it's a good thing that a geographer is bringing his message to the rest of the community. And I would have to say that geographers have a respectable position within the transport community because we tend to have a, that spatial view, the organizational view, the logistics view. And we defer, of course, then to colleagues who study supply chains in detail, who will have much more intimate knowledge of pricing and structures in those cases. But I think it's a fair case to be made that geographers have a strong position in locational analysis and transport, and what I would call operations research, which, you might think, Well, this guy sounds like he's doing logistics as a modeler. And definitely I am, so, somewhat a little bit out of the core of geography, but enough in it to be highly respectful. As I mentioned, my forebears within the discipline here in geography, this has been going on for a century, so people have this as a very solid aspect of geographical inquiry. David Staley 23:52 Tell us about your role as Divisional Dean for Social and Behavioral Sciences, what's involved in that? Morton O'Kelly 23:59 Well, David Staley 23:59 What's your portfolio? Morton O'Kelly 24:00 I joined the college last summer, and I've been happy now to have a year go by and to understand some of the cadences of the life of the college. I have really two roles. One would be to work, obviously, with my division, my eight departments of Social and Behavioral Science, and we work to advance all those great departments. David Staley 24:18 And what are those eight? Morton O'Kelly 24:19 We go through anthropology, communication, economics, and geography, political science, psychology, speech, and hearing science, and sociology. And so these eight are among the eight, I think, frankly, the best departments in the college, and they are tremendously productive scholars. I think to my benefit, having served as chair of one of those departments, I would not really want to be able to come into this position without having at least some experience in looking at from a chairs perspective. So advancing those departments is my daily job as well, the Executive Dean asked me to serve as the research Dean for the whole college and in that respect, I've had to broaden my horizon. I've walked to go visit with colleagues in physics, and chemistry, biological sciences, and meet them in their labs and learn more about what they do in order to be able to advance their cause in terms of the college and the research mission. So, it may sound a little odd for a geographer, social scientists to come in and be the research Dean, but it makes sense when you think that you can, you know, be just as adept at learning about psychology, philosophy, physics, and just to pick three very, very different departments, but all require some vision about advancing their excellence, and advancing their research productivity, and helping faculty colleagues do their best work. David Staley 25:41 When you teach students or any students who might be listening to this, what do you tell them about career prospects for students who major in Social Behavior or indeed the Arts and Sciences? Morton O'Kelly 25:51 Well, I think the great advantage of training in the arts and sciences and in social behavioral sciences would be that students get an ability to be generalists. And I think we've all heard that students these days are inevitably going to be having perhaps multiple careers throughout their life. Even if they're going to a professional school or going to get advanced degrees, the background they get as a good training in Arts and Sciences sets them up very well to be adaptable. Now, that's, you know, motherhood and apple pie, it's a broad range issue. But, I would say as well, that in certain fields, and I picked geography as one, there are also highly relevant technical skills that students generate and can learn about whether this be software it'd be use of systems such as GIS, geographical information systems, allow them, and by the way, in my research, much of the work I did was helped by having software and programming skill that allowed me to use data analysis to get the end product that I wish to obtain. I think students get that same richness, they may learn critical thinking, they may learn how to write a well crafted argument, they may learn how to produce videos and blogs, and they may learn how to know the most up to date techniques and skills from their scientific and arts background. And this is something that I think I'm very proud of in the college. David Staley 27:15 What's next for your research? I mean, I understand you're divisional Dean, that makes it harder to research, but you must have thoughts about what's next. Morton O'Kelly 27:22 Absolutely, and this is one of the things that gives me some joy in that I'm able to have a little time to continue my work on research. And actually, I have a fascinating project right now, where we got hold of hundreds of airline networks, I mentioned airlines before. So we have, now imagine not just two, but, hundreds of networks. All of these networks involve cities, involve connections, involve linkages. And, in some ways, it's like a global web of connectivity between everywhere from Tokyo, to Sydney, Australia, to London, to Hong Kong, to New York. Any individual airline only has a piece of this, but when you assemble them into an ensemble, that network has some very interesting properties. You might have heard the term the small world property, that is that many places are within a relatively small number of hops from one another. That's not true of an individual network, but when you amalgamate them into one big ensemble, it certainly is true. And I've then begun to at least think about and collaborate with other folks to see if we could find some new measures to make it easy to detect when a network is organized as a hub and spoke network like Delta, or when it's organized more like a point to point network like Southwest. And we in fact, in the world, we have hundreds of examples of both types that there's no one, no one size fits all here. And that's the wonderful thing about networks, there's literally dozens of types of organization. But primarily, they're going to roll down into maybe one or two big ones, then, you know, a central hub with many nodes coming into it, that node is going to have, what we call high connectivity. So which networks have that, which networks are more like say Ryanair or EasyJet, or Southwest to name three, that use a more point to point framework. And, I would like to be able to say, well, I could go through all of those 100 networks one by one manually and detect, what I really liked would be to write some machine learning code or some method to scan these data and tell me which ones typically fall within which categories. David Staley 29:24 Morton O'Kelly, thank you. Morton O'Kelly 29:26 Thanks very much. Eva Dale 29:27 Voices from the Arts and Sciences is produced and recorded at The Ohio State University College of Arts and Sciences Technology Services Studio. Sound engineering by Paul Kotheimer, produced by Doug Dangler. I'm Eva Dale. Transcribed by https://otter.ai