 Magandang araw mga kababayan at welcome to TVUP. This is the Science Innovation Series and today our topic is traffic and how big data might be able to help us solve our traffic problem. Our citizens are all concerned about the traffic. This is the walking public, the pedestrians, the commuters. Those with private vehicles and those who transport cargo through our major thoroughfares. I'm Giselle Koncepcion, professor at the Marine Science Institute of UP, Dileman. And I have with me my co-host Benji Valiejo, professor at the Institute of Environmental Science and Meteorology. Benji? Good day to our viewers and thank you again for having me here. We have with us two of our top young professors and scientists at the National Institute of Physics. We have Professor May Lim. May May? Hello po. And professor Giovanni Gani-Tapa. Magandang araw mga posay din. And they are complexity science researchers. But though that's a very abstract and computational modeling field, their concerns are very down to earth. And May works on traffic as an academic scientific problem, which we think is the way to try to improve the traffic situation in our country. So May, can you please tell us about what you're doing about traffic? So when we talk about traffic from my team's perspective, we talk about it in several ways. One is from a modeling perspective. Sinusubukan namin, pag ganto yung iniisip mo na mga tao. Ano mga yun yung iyari kong hindi lang siya isang tao, marami na kayo ang magahalu-halo doon? Ano mga yun yari? Pangalua, we also look into data. Pwede ba namin ma-validate yung model, mag-check kong tama nga ba yung prediction? Or the other way is we look at data and say, are we gaining some insights from that and saying, okay, if that's what data looks like, can we try to extend it into a model? And so this feedback loop of looking at data, modeling, and then trying to make predictions is essentially what we try to do when it comes to doing complex science-based analysis of traffic. So may do we have data, or for that matter, big data, meron na ba? Well, there's opportunistic data, but not any big data in the sense that I want, for example, this resolution, let's say per second or every 15 seconds, I would like to know where the buses are, that's hard. But there's opportunistic data in the sense that Google Maps, for example, would show you the volume of traffic in a certain given segment. MMDA posts and update every 15 minutes for certain points in Metro Manila. So those are the opportunistic data that's available right now. But if you want better data, there's data, but it's not in our hands. OK, so I think Benji has a lot of questions. Well, as somebody so affected by traffic, so. Yeah, well, my question for me would be this. When you say traffic, are we referring to people, individual people, or the vehicles? Oh, when we do some of our models, we look at vehicles, but we also look into people. Because there's also pedestrian traffic. And as we move on, because we've started with essentially what look like toy models, marang laruan lang siya na, OK, may kochika na gandto, susunod na gandto, anong mga iyari. Even those toy models, they give a lot of insights. But then we also know that, for example, we have models wherein we have people, we would say, OK, these are the people you say move. What would they do? How would they approach at a certain door, for example? And this is actually done with Gany a long time ago. Yeah, so. So there are those kinds of models. And so when we talk about traffic, we actually want to talk about everyone. We want to talk about movement, not just of people. We want to talk about movement of people at different scales, walking from one end of, let's say, katipuna to the other. That's too much, actually. But for example, just walking point A to B, even from the parking lot to your destination, it's part of the walking experience. Then we say movement of cars, and then the movement of cargo, and movement of goods. So when you say traffic, it should actually be a huge idea that traffic's about moving at a movement of people, goods, services. So may the model is only as good as the data, right? So maybe we can ask Gany about gathering the data objectively, not subjectively. And also earlier, we were talking about your initiatives in NIP about how to ease the traffic from your vehicles in NIP. Yes, ma'am. As May has said, we were working on this when we were graduate students. It really started with toy models. But the real problem is it's not really just traffic. It's at the whole question of transporting things. So it's a transportation issue. And the problem with this is, of course, scale. I have some students who are working, for example, in just looking at the pedestrian traffic inside UP. So how long will it take from a student from NIP to go to CHK, which is the gym? Typically, it will take, well, for me, that would be around 20 minutes. The least amount of time would be around 12 minutes. So we already have measured that. And it's a very simple thing to ask students to log it in their phones and then just report it at some point. So I asked several students, around 100 students already, to do this for me and log their point A, point B, point C, and the times at which they went. But this is not very easy to do in a scale larger than UP because then you would have to ask several millions of pedestrians all around the city, as well as the cars, et cetera. Ways has been doing this by automating, by using your cell phone. However, it only looks at the travel times. It doesn't follow you when you go into your classroom. It does not follow you when you go from one building to another when you're walking. It only works when you're in a car. It assumes that you're in a car. Now, on the other hand, we can build instruments, for example, a camera, where you can use this camera data or image data and extract traffic from this. I think Dr. Limas project like that. We can also use Bluetooth, for example, in our phones. And we actually tested this with UP Baguio, actually, in the Panagbenga Festival. So we asked some of our volunteers to actually merge with the whole Panagbenga crowd. And we were tracking their cell phones just to see if we can actually track in bulk a whole crowd moving in a narrow street. We were projecting to do this in the Nazareno Festival, which is a very hard problem to do. However, we started something small like that in Baguio. These things are available. These things we can do already. But as May said, there's data and there's big data. And we have to find a way to extract the most information from the amount of data that we already have, several students walking around in campus, several phones with Bluetooth in a crowd. Then we can now try to see if, using those, we can validate models that are available or predict data. My question on my student is, kapag lumabas ba ako dito sa NIP at tining ng ko ang traffic sa Katipunan, traffic ba mamaya pagdating pa sa Makati? And he tried to work on this problem. And he found out, well, at least he can predict 68% most of the time, the traffic situation when and I arrive in Makati. Of course, it's not always 100%. But 68% is better than guessing 50-50%. And I think there are a lot of improvements that can be done with the model of that student if we had better data. We were just using the MMDA data set. Yeah, I have a question for both of you. One of the malaking problema natin is, especially for travelers, to how to get to the airport on time. So I think for our viewers, this would be, some of our viewers are overseas. This is a very important piece of application. How to get to the airport on time? So if you get to Naia, how long will it take me to go to get to my house, let's say in Bulkan? So how could your models or studies give advice to our travelers? So they, and also to the airport authorities, what would be the probability or the chance that people will be late for their flights, for example. That's actually a very great question. I think we don't need to make a model right now for that. That can be answered right away because right now, for example, Google Maps would offer your Google traffic scenario and give you an estimate of how long it would take to go from point A to point B. And the quick answer, by the way, to that is, travel at 4 a.m. Between 2 a.m. and 4 a.m. 12 midnight to 4 a.m. That will be a consistent number. We've looked at it. It's the best time to travel to the airport, from the airport to the airport. On the other hand, avoid Wednesdays. Actually, it's interesting. We thought Mondays would be the worst time to travel, let's say, even along Edza. But it turns out it's Wednesdays. And this work, we've actually started this just two weeks ago when there was this bruha-habag. Okay, let's keep the trans-DNVS unnamed. But it was one of those things that we asked to traffic improve because they were suspended. And it was, it's hard to answer that. But the interesting thing is, because of that question, we were led to the ID to figure out we gathered data for about two weeks now. And it's consistent. Wednesdays are bad to travel, bad days to travel. Best time, no traffic. It's really midnight to 4 a.m. And it's actually true that on Edza, there's really not much, the concept of a rush hour is pretty gone already. It's like, it goes up and hardly goes down No let up, no reprieve. Yes, no reprieve at all. So to add that question, to from the airport, fairly, I mean, if you have to travel now, tomorrow, Google can help you with that. But you know, the problem really is, I think the question, do we really need the airport here? Or, as I said, it's a transportation question. It will limit our questions to just the travel, throughput time, the throughput or travel time. Then we will just be looking at it in the context of optimizing this for whatever parameter you want. Shorter times, shorter distances, larger volumes, et cetera. But there's another side of that. The fact that the airport is here right in the middle of the city, where we can actually move it somewhere else. And have a high speed train line. So these things has to be factored in if you want to really solve the problem of traffic. Data can only tell you so much. But policy really has to feed on whatever the data sets can tell you. The sad problem is sometimes the projects, the project lifetimes, planning lifetimes of projects is several decades already. And when they get to be implemented, the data where they were based on is probably not real anymore. And for example, in the U.S., they actually have this problem when they planned all their freeways based on 1950 data where the automobile was actually rising up. And then suddenly now, since 2004, there's a decline of automobile usage by the so-called millennials. And now you have too many freeways, too little public utilities, et cetera. And therefore, it's a problem that they based on data under 1950s. Now, I don't know about the projects here in Metro Manila and in other areas in the country. But we have to review actually all of these projects that are being proposed, especially in the light of the build, build, build directions of the current administration. They will have a very intense building of infrastructure that we... It might really ease it, the volume problem. It might ease it for a while. But there's a general rule actually in transportation. If you build it, they will come. In other words, if you expand your lanes to several lanes, make it eight lanes, cars will actually fill it up and your travel time will not be any better than the rest, except for a short, very short transitory phase. And so that's a problem. We have to look at it not only in the throughput time, travel time perspective, but also at what is actually the correct way to approach a problem of transportation. So that's a very important point, Gani. This is complex dynamic system. It's a very important frontier field of research in universities and even in the private and government sectors. And traffic is symbolic of everything else that the government would like to address to benefit itself, the governance as well as the private sector and also the individuals. So, Gani, you mentioned airport and road networks. And we should tell the public that presently there is going to be an airport that will be fully functional and the plan is to build the speed train. So that will decongest air traffic as well as road traffic. But even so, I experience already a great improvement in time going to the airport only because of the Skyway. Now that connects with... Is it E expressway? Just one intervention is good. Now we know that the government started in the last administration is also building this huge overpass and we're now suffering from the traffic it's causing. But it's going to connect north and south and it's got exits about eight to 12 of them within Metro Manila. And earlier, Benji was saying but that's another problem because the exits could be blocked. But I think it's important to tell our viewers that the government is trying to do something about it. But given the existing infrastructure and because we know that in the university we want to preach by example or teach by example, NIP is starting it with their own little interventions and rules because they know it's right. So a certain system, a certain kind of order in the way we move or transport ourselves and things is important. Do you think Benji that we could expand this to the National Science Complex, the College of Science and the whole UP, Diliman community, you are the environmentalist ecologist. Do you think that a reward system would be good for the community students, faculty, staff, a little reward to get the right data to contribute to the big data using maybe some apps that you would develop and using some instruments or sensors or what, cell phones. That would deliver us to us objective or non-subjective data. So Benji. Actually someone has made a study on that at one of their grad students at the Geography Department at the College of Social Science and Philosophy. She, Trina Listanco, she did a study on how students in UP Diliman perceived the way by which they moved from their class to the next class. And this was at the age before the smartphone. It was done about 10 years ago. So what happened is what resulted was that the students moved from point A to point B and there seems to be some gender differences on how they move. So that came out in her study. I don't know if that's the same today. So probably students are really willing to volunteer because she had a sample size of 300 students and they all volunteered for it but that's one of the more interesting results. Women tend to move in certain places while the men tend to move in certain places to get to the same class. One of the problems that we actually find when we try to use instrumented tracking is the amount of information you actually get. And one problem with this is that they don't really want to be tracked all the time and therefore we limit the tracking only at the times when they exit the building and to the next point that they come. This is a very simple non-electronic solution. We actually give them cards, yellow or orange cards and then they just submit it to a drop box in the end. And that's a very simple way to track them without being so intrusive about it. We actually do have developed in very intrusive code and code tracking devices but we use this for, well, we plan to use this for cats in the university because we actually have a feline problem in the university but the solution is not to take them out. The solution is to find out their ecology, how they move and then plan, just like the traffic we're doing. So these are very small devices that we can track for a week or so and attach it to the back of the animal. These ones we can also use for volunteers and some of them actually volunteers, volunteered to use this, but of course not all. But if you make it very simple and very just to give them a small nutrenumeration, just a recognition that they participated in this project. See sometimes enough for them to actually track for several weeks. So we actually have several weeks of continuous data for students who volunteer even when they went out to their dorm, they went out to their date, et cetera. Of course we anonymize this when we use this. It's a privacy issue. But once they, you get to get to understand the individual when they move. And I think that's another problem of this, another aspect of this problem of traffic. How the individual actually perceives moving from point A to point B. And I think the work of Trina is also important in that regard. So just to follow up on that. So the issue of privacy is actually a very big deal. I mean, on one end we say we want to have crowdsourced data, but then that could be very intrusive to someone. And on the other extreme, we have also this mindset. Why do we need the data? Could we just empower the user? And so with that mindset, what we would do is create the apps that would allow them to download the app and look at it. And so their own data is just theirs and they look at the things that we can normally calculate and they could evaluate what we normally evaluate for them. And with the hope that by having that information, an end user would make certain decisions to change behavior. And that is, of course, as I was saying, it's trying to change certain mindsets and that's one aspect we tried to do as a way. And it was very, very important. You started with the statement changing mindsets and that's a noble goal of training and education in the university. So there seems to be some ordering principles that are obvious to all of us and so they could be the basis for guidelines on how to deal with traffic, not just as an individual but as a community. Because on the other hand, if you want to get an individual to use an app to decide for himself, he will decide in an individualistic way. And then when the time comes, so it's a competitive way too. And the time comes when the machines are going to make the decisions for us based on the analysis of this. Complete data of the surroundings. Then the machines will, the artificial intelligences or the machine intelligences will kind of tell us what to do. That's another major topic for this science innovation series. But for the moment, we have to think of how we would manage traffic in this area. Say, UP Dilliman is a big enough community. And there are elders, there are seniors, there are middle-aged ones and the young ones. There are also children. And then you have a highly educated community that can be empowered, can be incentivized to collect the data, some level of data, as a basis for the studies that we're doing right now. Unfortunately or unfortunately happened during the transition of the roadworks here in UP Dilliman. So for the viewers out there, sometime, a few months ago, several roads in UP were actually closed and the usual routes of the jeeps were changed. So we had the data a few months of data before the roadworks and now we have the data after the roadworks. So I'm looking at it in a more positive light. So now I can tell you about how the community actually changed, responded to these change, how people would change their moving patterns. So the time cost from point A to point B has obviously changed when you write the eco-cheap which is the internal transportation inside the UP when it's now go and passes here in CP Garcia instead of the usual route in fine arts. So how do students actually respond? How do the communities in Christian Regas and actually do they still ride the jeepney? Because it will take you, a penalty would be around 10 minutes added the time on your trip and it would be interesting actually to see whether it will come back to the original pattern. Hopefully the roadworks will finish in a few months again so that we can still see what will happen afterwards. When we track the, when we continue on tracking the pedestrians even after the project has actually finished. There are very two important points that were mentioned just now and I just want to highlight them because they're great. One is that solutions are actually very local. Empirical solutions to traffic are empirical and local. Very important. And the second is the role of data is not just about collecting them, it's not an end goal. It's a way for us to assess whether solutions work or whether there's a problem to begin with. So that's what we try to use our data for not just to end that okay we have the data and those are very important that solutions are very empirical, very local and in nature. So I think it's important to emphasize the point that rather than just think of what the government should do for us in terms of the infrastructure or the rules and regulations, we have to think of how we can help solve the problem ourselves and when we say ourselves that includes the private sector and we also know that the private sector while contributing significantly to the traffic because there is unabetted development of say condominium complexes along EDSA. Then there's no way for the MRT to support this kind of a population movement. As you say, no matter how many roads we build, if you build them, they will come. So there must be a government macro plan or national plan for creating what we call hubs and spokes. And I also know that the government in the central Luzon region is going to build the New Clark City which is going to be a chartered city. So they're going to control the way it's being planned and they're going to make it a smart city and it's going to be smart for traffic. But the private sector, we're talking about the big, big investors, they're also trying to help build what you call live, work, and play communities. Just to try to reduce the traffic. Of course, you still have to go out of town because it's boring always to go to the mall, do the shopping and the eating in your place of work in your residence. But think again about what we have in UP communities, biggest of which is UP Dileman. But you have the whole UP system with eight constituent universities each with its unique ecology, unique traffic level. So then there's the surroundings. UP Dileman has the Katipunan Loyola area and then we have good neighbors like Ateneo and Meriam who have the same interests as us to ease the traffic. So we could actually expand whatever we started in NIP and then UP Dileman to this whole community. And you had told me also that this SM North is pretty well managed. So that's all within our community. It could become a model. So Benji, I think at this point you might have some thoughts on how we can empower this community to try to change the mindset, to follow certain rules about lifestyle, about schedules, about walking when you don't have to take a vehicle or do carpooling, things like this. Because you also lived and worked in Ateneo, right? That's so nice. Not really. What I used to teach there for a semester. So you know their mindsets. Well, I think one of the solutions and I spoke with Paolo Alcazar and the landscape architect who's going to give a talk to one of my classes in October in the Science Technology Society. He advocates that all planning should be local and everything should be walkable. So for instance, here in this area, Dileman, actually Dileman Miriam, the commercial center of town center, even Techno Hub. And if you're really for fitness, even SM North, they're walkable from Dileman. Yeah, they are. The thing is... Live, work and play now. How can we make people walk? That's the... Now, as May told me earlier before the show, mall owners can really make their mall goers walk along the mall. Now, there should be some sort of... Oh, walking regimen, ah? No, no. They go all over the place. So anyway, I think what we... Everything should be walkable. And UPD Dileman, I think has assets to make natural assets like the trees, especially the acacia trees, to make the whole campus walkable. But based on studies that may my students in STS, one of the major factors that discourage them from walking is not really that the climate is hot or it's raining. It's because the sidewalks aren't really built for walking. And I think this is true for a lot of the municipalities in the country, like architect Al-Khazavin told the audience at the National Museum two weeks ago. One of the things that we actually did as well is to extend the study to UPD. So I ask a colleague, Dr. Ranjabel Rojas, she is now collecting data as well on UPLB because we cannot really look at travel times only in... Well, at least for UP, only for UP Dileman and assume it will be the same for the rest. And UPLB has this Kaliwa Kanan roots for their eco... In their own jeep knees, the one that goes to forestry as a different root from the rest. And therefore it has its own dynamics. And we want to see what is the same and what is different from both campuses. And I think walking as well, it's not just the jeep knees since it's a pedestrian study. We also want to see what makes them walk more in UPLB and then in Dileman because we want to see... Well, because the end of the goal of the project is to actually propose a route for the eco, route for the talky, route for walking that is synchronized with the schedules of the students. Because sometimes the eco jeep would actually wait for a very long time but no students are coming in, coming out of the building because it's not yet dismissal time. And then when everybody comes out, 300 or 400 plus in NIP, for example, there's a very few jeep knees that are waiting for them. So these things we really have to deal not only in terms of traffic but also in planning for how can you make the rest of those who did not get a ride walk very fast so that they can meet and not be late in the next class. So I know it's really human behavior and lifestyle that is a major determinant of how efficient our transport or our traffic movements would be. And I can think of one good reason to walk and that's for health and wellness and it's got to be brisk walking but then our classes should start maybe five minutes after the break so that people have the time to walk. Now, we know that in UP Diliman, Chancellor Michael Tan had this plan to build a bike lane and we already had this plan to come up with rentable bikes that can be secured. So I think that should be pursued but for the pedestrians, the ones who really love to walk briskly, we need covered courts. So UP Diliman could really be a model for other academic communities and once eventually, urban and rural communities. So then again, there are those who cannot avoid traveling to another location, a distant location, say between UP Diliman and LBO between Quezon City and Makati. So we already know that, you know, operations like Uber or Grab, together with Waze and Google Maps have influenced the way we managed transport and traffic. So we would like just a brief comment from May and Gany regarding this. We're sort of running out of time and there's just so much for us to discuss but I think it's important to tell the public how this is affecting us positively. Okay? Or not? Well, for one thing, I think that the NBS, like Uber, Grab and all the rest, this actually makes for users aware of situations outside of what they can see and therefore plan their route effectively. Of course, that's on top of the data that Uber and the rest are gathering. But I would like actually to have a more integrated solution. That's just one that is left only to Uber or left only to Grab, et cetera. I don't know if they would share data, probably not, but the MMDA can actually feed into their data and I think they have memorandum of agreement with them but the MMDA should also tap actually studies that we can do, for example, here in the university. On the other hand, I'd just like to add for a final comment that we also have to exploit other areas, for example, the waterways. We have a very wide waterway. Of course, everybody does not like Pasig River because of the smell, et cetera. But it's not the smell. And it's already a very, it's a river that cuts across the whole Metro Manila, the whole Metro. And as well as that you also have the Naguna Lake system, it would take you three to four hours to go to Baras. But if you actually have a ferry that could bring you there, we can decongest along the city by bringing other people away, not really away, but make them accessible to the Metro and not stay here for the rest. So that maximizing whatever we have is actually very important. So, May, yes. I'm sure you have lots to say. Well, for me, companies like Uber and Grab, they serve a very, a nice purpose, which is they're helping potential car owners not own a car. I mean, I'm one, for example. I like their services. I use their service. It's because I don't mean, I don't want to think about parking. I don't want to think about parking at home or even where I'm going. So they do serve a purpose. But at the same time, we always should also see the other side of the coin we're in. One more Uber, one more Grab vehicle. Might be too much, but only if there's an excess in supply versus demand and as a result, they're going to be stopped somewhere and therefore they're going to be part of the problem. But it's a balancing act. And having, and it seems that economics would guide the principle whether those cars would disappear from the system altogether. But there's also some form of regulation that's also being done, which, for example, in terms of data, they could track how long the response time between hailing an Uber and getting one would take or how long a driver actually waits for someone to hail him. So these are things that could be data-driven and these are things, and they do serve a purpose. On the other extreme, when we talk about, let's say, Google and Waze, they do have some exchange of data with local governments. So it's just a matter of local governments being able to tap those sources. And I almost forgot, Uber actually has an initiative called Uber Movement wherein they share some form of aggregated anonymized data that tells you the travel time between points A and B. And their data is actually open source and accessible. But again, the level of anonymization that they've already done and aggregation that they've already done might not be sufficient for our purposes already. So there are ways to do this. And we actually are going to start to try to reach out to them soon. There are ways to improve this. May, yesterday, I was on my way back from Los Baños and I was asked to pass through C5 and then Temple Drive to Edisa. And I wondered why? And so we decided to take C5 all the way to Blue Ridge and Catipuna and then there was no traffic. So there are ways to improve these monitoring apps. And well, I'm happy to say that NIP and our engineers from College of Engineering are working on these with major funding from our government. But Gany is a leader of a Vicer program that is funded by the University of the Philippines, stands for Versatile Instrumentation for Science Education Research. So there are ways for Gany's team to come up with monitoring sensors that could help problems such as traffic. But I think, Benji, you've got to have the last say on this because you are the ecologist. So, you know, what's the game plan for UP? For UP? Well, I suppose we have to make the campus, well, in my opinion, should be walkable for all, promote transportation that doesn't emit greenhouse gas. And I think there should be a way to regulate the entry of vehicles because say, university campus, it's not a commercial district. So I think there must be a way. And I think the university has started by having a sticker scheme. If you don't have a sticker, you can only pass certain routes. Precisely, yeah. But the university, I think it's a good model. It's a, UP Delemen is as big as a small city. So whatever we come out here, may be applicable to other cities, especially regional cities of the Philippines like Iluilo, for instance, where a lot of the development is really positive. So, yeah, I think I would have to agree with that, Benji, that there should be some rules, I wouldn't say dictated, but strongly implemented from the top because they're obviously the right rules. And it's really based on your good ends when your fellow neighbors good begin. So there's always this balancing. And so, I think stickers follow the rules or don't be a pollution belcher, those things are important because ultimately it's the individuals who must live healthy, productive lives, long lives in the university and in the rest of a Philippine community. So with this very rich, stimulating discussion on traffic, which affects our everyday lives, we hope that our viewers would keep tuning in to science innovations in TV UP. No Way But Up, TV UP. Maraming salamat po sa inyong lahat. Magandang araw.