Designs on the Games
Tuesday 7th October 2008
Savoy Place, London
Organised by PhoneAbility and the IET
Travel information and special events
by Prof. Mike McDonald
Good morning. What I am going to do this morning is talk a little bit about travel information more generally, some of the means of delivery and then make one or two comments as it relates to special events particularly with a focus on the Olympics. We did have an involvement in some of the modelling work for the Beijing Olympics.
If we look at the objective of travel information, its just really two. First one is to enable people to plan some way before a trip of theirs. There's also a requirement for information during the trip, to enable people to be confident that they are making the right journey, in the right sort of vehicle, and also particularly for journey recovery when something goes wrong. It can be quite a useful thing journey recovery
These two points are valid for individuals making decisions for themselves, for a group which may have a range of different users needs or for people who are making decisions on behalf of a company, for the movement of goods as well as for people.
What is required to meet these objectives? The first one is an accurate and reliable information base. The focus of information very often on the delivery mechanism, I will say something significant about that, but underlying that you have to have a very good understanding that you are able to deliver what is required for those people who are using the information.
You need to be able to deliver it both before, and during travel, a series of mechanisms for doing that. It has to be effective, the human machine interface has to be such that you can get access to information. You very often have difficulties if you are using the internet to get information, that you need a large number of key strokes to make a large number of decisions before you are able to focus on the information you are getting. It has to be delivered effectively. I can give an example of that of a Japanese student last year on the first day of term. The university is supposed to be good at delivering information, I don't think it is particularly, but it has big signs up which are about 8 feet by 8 feet which have the layout of the campus on.
This young lady asked me where a particular building was so I showed her the plan, and what she did was then she used her mobile phone to take a photograph of the plan. She then had an image which was about 2 centimetres and immediately wandered off in the wrong direction. Effective delivery of information in a way that people can use is important.
The last is getting a clear understanding of this. A lot has been said about that too. We're involved in a number of UK and European projects. The UK projects will tick the box of doing a user requirement study which means they focus on the norm, an engineer, as they are known as now, will focus on the norm rather than get the spread. Then there are individual projects which look at people, information for people with disabilities. They don't tend to come together in a coherent mainstream solution. There is a major problem remaining with that.
If we look at typical information systems, there are variable message signs, in vehicle information message systems, the internet, radio, TV, kiosks and telephone. What I will do is just look very briefly at one or two examples of these, to give an indication of where we are standing with those.
Firstly if we look at the road side variable message signs. Typically you require 3 pieces of information. You need to understand what the problem is, you need to understand where it is, and you need to understand something about what you should do about it.
So generally, variable message signs will and should have at least 3 lines of information within them. They are very useful, 67 % of drivers over a very large number of surveys find them useful.
In any event, such as the reporting of an incident ahead which might be a break down or a problem on the road typically, maybe 10% of people, it's a relatively small number, will be aware of the information, but they will wait until they see the problem themselves before they choose to do something. That is probably good news because if you look at a network, then if you have a problem on a key road, what happens is that people divert away from that and you get a cascade of congestion moving across the network. Very often on a major road the best activity you can do if you have a total road closure, is not tell anybody for a little while and park as many people as you can on that section of road. {Laughter}. So there is a problem here as to telling people information which is timely, correct and useful, and using the information that you have on the network in order to manage it most effectively to reduce the total hours of delay on the network. It met not be evenly distributed.
In an urban network, even if you have major problem on a link you would ought not to have 30 or 50% diverting otherwise you will have major network problems. This emphasises that you need a very, very good understanding of what is happening on the whole of the network, which you do not have in the UK, or anywhere else for that matter, and with that understanding we have to be able to model the impact of the effects if there is an event on the network.
If we look at bus stop variable message signs. These are very good. Bus stops, railway stations whatever, in telling people the next trains, next buses that are coming and destinations. Park and ride sites as well. The service has to be reliable, interesting from the surveys that on 2 or 3 surveys that the network the service itself has been less reliable after the introduction of variable message signs, but the passengers have perceived it has been more reliable because they have information that they didn't otherwise have. So information is very valuable within that.
There is a problem of consistency of human, machine interface.
Vehicle navigation systems are increasingly used and available within ... generally the network information which is available is very good. We hear horror stories of heavy vehicles using roads which are totally unsuitable, and there is an issue that part of the network is environmentally sensitive, as to how you manage people out of that network. The only way of doing that is to put some constraint on the network as a physical constraint, which may be a speed, access whatever. Because the information on the network is out people will use it.
One issue is whether you have one-way or two-way communication with the operator. If you have two-way communication, increasingly you can take the information from the users and use that which gives you online network journey times for the people. Japan is way ahead of us within this, one of the reasons for that is that they set up a private company probably about 15 years ago now, to generate a digital database of the network which enabled the operators then to provide information on traffic flow. That is probably what has delayed the implementation of in-vehicle technology in the UK, why we had so many imperfections in the system and services now.
Increasingly there is information provided within the vehicle, that gives people confidence that they are in the right vehicle, and that they are going to right destination. Generally that information is visual only, and that is an issue for those who are visually impaired. For the roadside information, for the information at bus stops and railway stations. You can get systems for visually impaired, where, when they get to bus stop they can communicate and they will then get a verbal description of what is on the variable message sign. Its not that straight forward, and certainly for some of the applications some of the local residents are complaining about the noise of hearing the instructions about which buses are coming next. So there are issues about how you produce information.
The internet is another application, and there are several internet services, www.transportdirect.info is probably the UK's leading edge on this with multi-information, it also provides information on carbon usage for any particular journey that you are planning, and although that is very general at the moment it is probably a significant step forward.
Recent surveys which we have done have shown that the majority of people are aware and agree that global warming is happening, they are aware that their journeys are contributing to it, but they will not change their travel behaviour because they don't feel that they as individuals are likely to make a significant affect.
But nonetheless that sort of information is there and available. You can get map information associated with that, and surveys have generally shown that people like map information. When we looked at variable message signs, road side in the UK, we generally have messages which are text messages. People generally don't like them. They would like a message which has got a map on it. There are problems because generally the way in which females look at maps is different from the way in which males look at maps. {Laughter} It's the orientation, whether it's north up or directional. One is not better than the other. If you have a road sign or message sign, then the one will see it better than another group and if you take the motorway ring round Birmingham, how would you set a variable message sign, would it be north up or directional up, going two ways. There are significant issues within this.
There are information signs which lead you to information signs which give you frequencies for radio, whatever is in, round variable ones. The first one went into America in 1940, related to the George Washington Bridge. That is a very simple very straight-forward way of using it because you have a very clear objective. One of the problems with road side spot location signs is that they only have a value for a significant relatively short way ahead. So if you are travelling 200 miles, a variable message sign is not going to be very much use because many of the problems maybe occur further up, away from your present position.
Traffic information kiosks, we have heard about and are not generally liked. The human machine interface is very difficult for many people to use, many of the surveys have watched people trying to use them and after some minutes they have walked away probably without getting the information that they were seeking within that. So, they are significant, but there is a great deal of work still to be done on those, and generally they don't work for the visually impaired.
SMS. Short message system alerts are becoming more common, as are telephone messages. In Turin for example you can sign up to a service and they will tell you if there's a problem on your regular route that you will use to travel to work. SMS messages are available in the same sort of way but generally, they don't give the breadth of the window of modes of timing that people want. So they have some limitations within that.
The last one that I will talk about within the usage is that of a handset. One of our studies at the moment is the application of mobile information systems. The aim is that the individuals will get a map on an advanced telephone screen. That map gives them their location and will guide them to their destination. Online information is fed in from the local authority, which gives online public transport information, location of bus stops and so on.
Generally the surveys show people really would like that and will appreciate that. There are a very large number of problems and issues associated with them. One problem is the location. If you are being guided to a bus stop you need to have a location accuracy which is greater than is generally available with GPS. Another problem is the routes. We did a survey of pedestrians some year or so ago, and they were tracked by GPS devices, and they were asked to go from a destination in an area they generally knew. A surprising result was that most people who travelled by car generally chose the routes they would have taken by car, even though they were 20, 30% longer than the pedestrian routes. So there are issues about interpretation.
If you have a pedestrian route, a pedestrian route during a working day may well take you into a shop, for example, you may enter Boots at one side and come out on the other side on a different street. How do you build that in? We have buildings in these routes where there is a problem with consistency and accuracy. It's a major way forward, it's a major beneficial activity that should have very good developments for those who are visually impaired and those who are disabled. Because you can put in crossing points where there are dropped curbs, you can put in a whole range of other things in here, but who pays for that database is not yet really clear. Who pays for the additional location accuracy is not yet clear. So the technology is ahead of the business case in that.
Key information issues, the quality of information, the static and the dynamic information on the delivery systems. Let's look at one issue. This is a slide which is a graph on it, which shows an estimated delay during an incident on a section of road and the forecast.
The estimated delay is the line which has the straight lines in it, and the other one is the variable line within that. We can see two aspects of that, one is that the delay is not necessarily long, this is some of the best alghorithms on the simplest part of the network. The second is there is a delay, a latency. The delay occurs - it's not until a while later that the clarity is there, it's not until some while later that incident is off. Any network whether its public or private the problem is forecasting sharply the start of the incident, and also at the end of the incident.
This requires high level hardware. For this particular part of the national motorway network, every link on the network has data measurement, loops which are cut in the road, and automatic number plate recognition cameras measuring number plates in two locations to get journey times. Collecting data is expensive, and the delay in application of most of these systems is due to the delay rather than due to lack of equipment and the delay in getting a knowledge of what is happening on the network compared with the delivery system.
So we look at the information process. Data collection comes from traffic management systems. For the Beijing Olympics they had a major new system in there with huge amount of money to collect and interpret the data within there. It comes from operators of a whole range of services - comes from service providers, buses whatever, comes from the users. A wide number of people will telephone. In Holland there is an emergency service system, one of the main reasons for that is that if there is an accident on the motorway, the police and ambulance services get totally swamped by the number of people ringing from mobile phones, and they require quite some time to understand the problem, by having an automatic emergency call system.
Collation and interpretation, we are many years away from getting a coherent understanding of what is happening at any time on most of our public and private networks. Sources some from a lot of different sources of data, and you need to understand the accuracy and the waiting for that. The management we need to make this a priority needs to be forecast, you need to model within that. You need at end of that to disseminate the information, and you need to control. And of course major activities for networks, we have modelling plans and any particular network it may be 2,000 or 8,000 different plans. If it depends where it is, how severe it is where it happens. You have you have to tell people, move them away from the problem and at the same time you have to change the level of capacity there.
Special events have a lot of different user groups: its participants, those spectators, emergency services, transport service providers, security, you have regular travellers using the network. All have different requirements.
If we look at the participants, if we're looking at something like the Games, they have to have an absolute priority. Not just that they arrive on time at the right place, but they have to arrive in time without having been flustered or worried or concerned that they will be late. I will come back to those examples in a moment.
The spectators need sufficient information to make earlier decisions. Again I will come back to some comments on that. Media and other groups and security services all need their own requirements within there. And regular users as well within there.
Let's just have a look at some of those. If we look at the Commonwealth Games in Manchester 2002, which are much smaller than the Olympics. About 80% of the people going to the Games went by public transport. For London, I think the target is 100%. The cheap fares and the online information system took about 200,000 car journeys from the network, it ended up saving almost a million car miles within there. The good thing was that at the end of it there was a legacy that there were more people using public transport who had got used to using it during the period of the Games that they stayed using it afterwards. Quite a significant legacy within there.
If we look at the Beijing Olympics, then typically every group was planned for with route information, which related to where they were going, why they were going, when they were going, the times, operating times of many services, journey times as we can see. Pick up points, drop off, access and entrance. Together with all the other package of security. So there were packages put together for every group and every individual. The number of people, of Chinese people, who supported the Games was enormous. So at any point somebody had an uncle or an aunt who would take care of them and move them forward if they were having problems. That was the plan and I think enlarged it generally worked within there.
That is not something that is likely to be available as we said earlier, in London, and therefore the technologies and the clarity must be much better at the start of the Games. It has to be accurate it has to be detailed and has to be credible. One of the things we have found from other Olympic venues is if the information, however good it might be, does not come across in a credible way, people will not use it. So credibility is quite important within there.
I won't say too much about the Olympics because in the UK there is a huge amount of work which is going on with that. But just two points. One is that provision of information is totally crucial to any transport service, before and during. What makes it credible. That service information has to be provided to all users. There are always going to be problems, some will be foreseen, some may not be foreseen. What is crucial is to have spare capacity. What they did in Beijing was to have special lanes for Olympics, they took the major part of the network and defined it as the Olympic road network, that was managed with policeman at every access point, and every exit point and on ways of getting into that network. The network was managed to a level of traffic which was significantly below capacity, in order to allow some spare capacity when accidents or incidents or anything occurred within there. The degree of control offered is something that would not be available within the UK. So there is a major problem of how you create and deal with capacity.
There are always going to be problems. Some will be foreseen. Some may not be foreseen. What is crucial is to have spare capacity.
What they did in Beijing was have special lanes for Olympics. They took the major part of the network and defined it as the Olympic road network. That was managed with policemen on every access point and every exit point and on ways of getting into that network. The network was managed to a level of traffic which was significantly below capacity in order to allow some spare capacity when accidents or incidents or anything occurred within that. The degree of control over that is something that would not be available within the UK. So there is a major problem of how you create and deal with capacity.
The information is only good if people can use it. There have been several park and ride sites where information systems have been put in where people approach them which says, "Please use the park and ride. The next bus or tram will be in 10 or 15 minutes. And then, when they have put journey times on there, some of them will say, "If you use park and ride, your journey time will be 45 minutes. If you don't, it's 25 minutes by car."
So it has to be valuable information within that. For Strasbourg in Europe, we did some surveys. They had park and ride sites. They had environmental quality measurements, air quality measurements in the centre. Where the air quality was poor, they gave that information and people approaching this main park and ride site to the city said, "Air quality is poor. Please use public transport." They got about 30% of people using public transport when those messages came up.
So people can be sensible if they get the right messages. But underlying capacity is important. There are different ways of achieving capacity.
As you can see on the film, we have an underground system with a large number of passengers waiting. There clearly is not enough room for everybody! (Laughter) Obviously some people will be left behind. It's just not quite possible to get that number of people into those coaches...
I have a feeling the doors at the other side of the coaches are probably opened with people pushing them in! There we go... A job well done!
They were clearly not the red-carpet people! Thank you.
(Applause)
JOHN GILL: We've got time for one brief question for Professor McDonald. Tony Shipley?
Tony Shipley: I fully take your point that the information that's presented says the car driver must be accurate and up-to-date.
I think, from personal experience, there is another element which is important, and that is that the recipient must actually believe that information. It's a confidence issue.
For example, I have experienced more than once situations where the sat nav has said, "Turn left" and I thought, "Nonsense. It doesn't know the way, I do." And then two miles further on I ran into roadworks.
So it did know. But how do I know that it knew?
A. It is a question of confidence. If there is a major problem in a network, then your journey is always going to be worse than it would otherwise have been, and the difficulty is knowing whether it would have been even worse if you had not had that information.
It is simply a question of confidence, and you will talk to others who have travelled during the day; you will have other feedback and so on.
But if you feel you have been given the wrong information, the evidence would suggest that that has eight times the influence in making your decisions than if you have been given the wrong information. So it has a very substantial effect on the credibility of any system within that.
If you are looking at something like the Olympics, then you may have people who are staying for two weeks or they may be staying for a longer period than that.
If on the first day it is in any way a shambles, that will affect decisions throughout the whole of the rest of the period.
What they did in Beijing was they set up the systems and they ran them for a significant period before the start of the Olympics within there, so that that would not occur.
JOHN GILL: Thank you very much. We now move on to Geoff Doggett from SmartAngles. Geoff is also the chair of the Smartcard Networking Forum. Mr Doggett.
| Previous |
Last updated: 20.11.2009 © Copyright reserved Website design: Digital Accessibility Team