How would it work?

"I had a great idea this morning, but I didn't like it." (Hollywood film producer Samuel Goldwyn 1882-1974)

We will devote this page to a short overview of how a completely automatic transportation system like the FLYWAYŪ would work. Those readers who want a more detailed description of the functionality should read about "the Addressing System" and other web-pages. These specifications are mostly our own. No doubt there will be other (maybe even better) ideas as to how this should be implemented. There are experts in Sweden and in the USA (probably in other countries as well) who are well versed in this subject. Generally, they each advocate different systems, not agreeing which one is the best. Some even favor the supported system. We will here describe how the FLYWAYŪ system, with a so-called "point-synchronous" network will handle traffic, in a metropolitan area where network has become fairly large and complex.	Guiding a car through the network How weaving affects speed Comparison with freeway traffic.

1) The central computer directs the vehicles used for regular traffic to their respective starting points, in accordance to their timetables, and taxicabs and freight vehicles in accordance to orders and previous bookings. The vehicles are informed about their destinations, and will find their way to their starting points in the same manner as when they are in service. CONTINUED ==>	Generally speaking, these empty vehicles will have a lower priority should the network be congested. In really tight situations, these vehicles could even be shunted off the trunklines to let other traffic pass through. If this situation is foreseen, these vehicles would be given a long enough headstart so that they would get to their starting points in time.
2) Changes in the trafficability of the beam net are continually reported by the nodes to the central computer, which in turn distributes this information to all other nodes (and maybe even directly to affected vehices). the nodes will uppdate their databases, and in passenger vehicles the travelers will be informed about delays, if there are any.	3) Whenever a vehicle leaves its place of departure, this will be reported to the central computer, both by the vehicle and by the first node that it passes together with information about its destination.
4) As soon as the vehicle leaves one node, it will inform the next upcoming node that it is approaching, and its destination. Should this node be a divergence node, the vehicle will receive information about recommended route, considering the traffic situation.	This information could consist of choice of route for several nodes up ahead at places with several shunts close together, such as in terminal areas. Should the node be a convergence node, the vehicle will receive a preliminary time of arrival to this shunt. The vehicle would then adjust its speed to accomodate this time.

5) When the vehicle passes the booking point, the node would be adviced about this both by the vehicle and by indicators at this point, inside the beam. Then the vehicle would be given a definitive time of arrival to the shunt. The vehicle has now been put into a timeslot, and it adjusts its speed to stay there. The main task of this node is to calculate timeslots for arriving vehicles (such as nr. 1 in figure 8), considering that vehicles might be arriving on the other adjoining branch (2 and 3). The traffic is "weaved" in together in the same manner as motorcars entering a freeway. Each adjoining beam thus have a booking point. One of them consists of closest foregoing node (node A in figure 8, since B is further away). The other booking point would be at an equal distance away from the shunt. The longer the distance from the shunt, the smooter the travel would be, as the speed adjustments would not have to be so sudden.	Figure 1:1
6) When the convergence point is passed, the vehicle will inform the next node that it is on its' way. Should this be a convergence node, this procedure would be repeated. On the average, 50 % of the booking points will coincide with the foregoing node (figure 8).	7) After passing the last node, the vehicle will inform the node of its destinationen that it is about to arrive. This would not always apply to taxicabs, however, they would be permitted to stop almost anywhere, as long as they are not unduly in the way. When the vehicle has arrived, it informs the central computer. This would be logged, and quite possibly the computer would assign a new journey to the vehicle. Should it not be immediately needed, it would be classified as an empty car and be directed to a depot somewhere.
8) The computers of the divergence nodes have a table of destinations for the entire beam network. This table contains information about which way an arriving vehicle with a certain destination should be shunted. This table is complemented by a similar table of a temporary nature. It is continually being updated by the central computer and reflects the generall traffic situation on the net, as reported by all the nodes. So, the first table, modified by the second table provides the shunting information being given to the approaching vehicle. The computers of the convergence nodes have the duty of fitting arriving traffic into timeslots, as described above. So, they send information about expected arrival time to the shunt. All nodes have to keep vehicles informed about any general speed limits on the next beam segment.	9) Each vehicle should, apart from this information, also use its scanning radar to maintain a safe distance to the vehicle up ahead. If an emergency halt should be necessary, the vehicle should afterwards back up about 10 meters. The purpose for this is not to unnecessarily load down particular segments of the beams with too many beam-vehicles bunched together. The total supply of seats and the type of vehicles available is decided by demand during rush hours. The central computer makes a correlation between the available prognostication and the actual traffic situation.

Shunting eaving is the process of joining the traffic of 2 beams together onto one beam, while divergence shunts forces an approaching car to choose left or right. Functionally, there are thus two kinds of shunts; for merging traffic and for diverging traffic, respectively. Divergence shunts are of 3 kinds: Those where both beams after the shunt have such sharp turns that the vehicle has to slow down ahead of shunting Those where vehicles that pick one of the beams can proceed without having to reduce speed Those where vehicles entering both beams can maintain their speed. These are described on the page about Shunts and Shunting. Naturally, the third kind is the most desirable, but it cannot be achieved in all situations. And weaving shunts are sometimes followed by sharp turns in the beam that forces a car to slow down in order to better handle centrifugal forces. Weaving Shunts Assuming that we have a weaving shunt where all beams involved are straight, and join up at a narrow angle, it is still not certain that vehicles entering the shunt can maintain their speed. This situation would, of course, occur if the feeder beams have such heavy traffic that the common beam (after the shunt) cannot swallow that traffic at the same speed. Look at figure 1:1 above. The booking points are always equi-distant from the shunt. So letīs assume a simple example: The feeder beams, A and B, have the same permitted speed as the common beam, C, where the traffic will join after the shunt. A bunch of cars, consisting of 10 vehicles, are approaching on beam A, and a similar bunch on beam B. They will both arrive at the shunt at the same time. With the term bunch, we mean that the 10 cars are grouped together so as to occupy 10 timeslots in succession. The booking points are so far removed from the shunt that all 10 timeslots fit into the distance between the booking points and the shunt. What happens now is that when car number one on both beams pass their booking points, the node computer will allow one of them (letīs say the one on beam A) to proceed with undiminished speed. (2 bookings are never processed at the same time. Even if they should be made at same split-second, which is highly unlikely, they cannot be processed simultaneously, because the computer can only handle one incoming interrupt at a time.)

The other car (the one on beam B) would be slowed somewhat, so that it would enter the next timeslot on beam C. How is that speed calculated? Well, assuming a permitted speed of v (in meters per second, to simplify calculations), and the distance from booking point to shunt is d meters, the first car would take d/v = T seconds to arrive in time to land into timeslot 1. If the timeslots are t seconds apart, the first car from beam B (letīs call it B:1) would arrive just in time for slot number 2 if it adjusted it speed so that d/v = T + t, which gives v = d/(T + t) m/sec. Car A:2 would normally arrive t seconds later, just in time for slot 2. But since slot 2 just got occupied, it would have to wait t seconds for slot 3, and its speed would consequently have to be v = d/(T +t) m/sec. Car B:2 would suffer a further delay; its speed would be v = d/(T + 2*t) m/sec., and so would the speed of car A:3, and so on. We find that the speeed of the last cars would be: For car A:10: v = d/(T + 9*t) m/sec. For car B:10: v = d/(T + 10*t) m/sec. This can be shown graphically as in figure 2:1, where the slope of each line is proportional to the speed of the car. Figure 2:1 As can be seen, the travel time T2 for car A:10 from booking point to shunt is considerably longer than travel time T1 for car A:1.

e can note a signicant difference between road traffic and automatic beam traffic when it comes to merging traffic. Two cars (B and C on figure 3:1 at right) are travelling on the freeway at a comfortable distance. When a new car (A) enters the freeway, he will crowd into that space between B and C. That is how it is supposed to work. This way, none of the vehicles involved should need to take immediate action to slow down. After A has entered the freeway, however, both A and C will have to reduce speed somewhat, in order to get back to the distance which is comfortable considering their cruising speed. Before this is done, the vihicles will, for a few seconds, be too close to each other for comfort. We can see from the foregoing chapter that with automatically controlled vehicles on a computer-supervised network, it is the other way around. The network can see vehicle A approaching, and can adjust the speeds of both A and C before the merger, so that required safety distances are maintained at all times. From this, one can see that merging of traffic in automatically controlled networks is actually safer than merging manually controlled traffic onto freeways.

Figure 3:1 Figure 3:2