Monday, August 11, 2008

Energy Efficiency of PRT - calculation

Someone asked me, "How can the PRT be more efficient than the MRT?" The MRT has economy of scale and should use less energy. I think that the PRT can be more efficient because:

1. the number of podcars can vary according to demand.
2. the podcars can go straight to the destination, without having to stop at the intermediate stops

A Transportation Enthusiast provided the following useful information for my calculations:

> Consider weight alone as a rough estimate of energy usage. This is not fully accurate, but can give a ballpark number for energy efficiency.
> According to Wikipedia, MRT trains weigh about 214,000 kg unloaded and can carry a maximum of 1920 passengers (492 sitting, 1428 standing).
> A ULTra PRT weighs about 700kg unloaded and carry 4 passengers.

I will approach the calculation in this way:

> Train 214,000 kg / 1920 passenger = 111 kg p passenger
> Podcar = 700 kg / 4 = 175 kg p passenger

Assume an average occupancy of 50% during the peak and off-peak hours.

> Train (50% occupancy) = 222 kg p passenger
> Podcar (50% occupancy) = 350 kg p passenger

We have to consider the energy used in starting and stopping the train. Assume that this adds to an additional 20% in energy use.

The final adjustment is for distance travelled. The podcar can take the direct route and cut down the distance. The train follows the same schedule and forces all passengers to travel longer. Assume that the difference in distance is 30%.

The final energy consumpiton is:

> Train 222 X 1.2 X 1.3 = 346 kg p passenger
> Podcar = 350 kg p passenger

Both train and podcars can probably consume the same energy.

The podcar beats the train in the following respects:

> Shorter travelling time, due to direct route
> All passengers in the podcar are seated.

If the podcar can be filled with 4 passengers (through pod-pooling, to reduce cost per passenger), the pod beats the train in energy consumption as well.

Disclaimer: This is just a simple calculation. It is based on assumptions that may differ in actual practice. Take the conclusion with a pinch of salt!


  1. The other advantage of PRT [or podcar] that hasn't been mentioned is the 'inefficiency' of passenger waiting times. In a prt [or pod car] system the vehicles are idle [and charging] while THEY wait for a customer, MRT the PASSENGERS wait for varying times depending on time of travel.

    At times of low demand these waiting times can be upto 1 hour or more and the occupancy can be as low as 1 or 2 passengers per carriage.

    The PRT is a totally automated 24/7 operation with virtually the same level of service at all times [infact the offpeak service is actually marginally better - making it, along with its privacy, a realistic replacement to the car, especially as you can use it when drunk without risking loosing your licence or livelyhood]

    The other efficiency issue that has been forgotten is the PRT has regenerative braking, and can be easily partially or completely solar pv powered once the pv prices start to fall into a more affordable price, [and or the electricity price goes through the roof]

  2. A Transportation EnthusiastAugust 12, 2008 at 6:32 AM

    I think 50% occupancy is very high for both. For PRT, the lower bound on occupancy is 12.5% but this is the absolute worst case for 4-passenger PRT, when every vehicle makes a full round trip taking a single passenger in the first leg and returning empty in the second leg. More typical is 1.2 per vehicle and 30% overhead for empty vehicle movement. So that's an occupancy average of 1.2/(4*1.3) ~= 23%.

    Train occupancy is much more complicated, because occupancy is inherently a function of schedules and demand (and, in fact, demand is also a function of schedules because fewer people will ride if schedules are cut). Also, demand and schedules vary over time of day and day of week, so it gets complicated quick and generally you need to rely on aggregate statistics.

    In my research on US statistics, I've found that it's very rare for trains to exceed 20% occupancy while maintaining a decent off-peak schedule, because every off-peak train is dragging the average down significantly. But you can't very well eliminate those routes, so they are a necessary evil.

    You should seek out MTA statistics for the final word. In the US, the numbers are reported as "total passenger miles" and "total vehicle miles"; you divide passenger miles by vehicle miles and that's your average passengers/vehicle. Then, if you know maximum passenger capacity of the trains you can estimate occupancy. This may be complicated if the train sizes vary from route to route, or over time.

    But I think it's a safe general assumption that 4-vehicle PRT will have higher average occupancy (~24%) than a train (10-20%), and even that is only because of optimized train schedules - this is an important point, because if trains ran at full service all the time (like PRT), train's numbers would drop down into the low single digits. So trains can only compete with PRT by slicing off-peak schedules.

    One other comment on taranga's comment: as far as I know, most current PRT systems do not have regenerative braking, for the simple reason that the amount of energy recovery is very small for the lightweight vehicles and infrequent stopping. In fact, the added weight of including regenerative braking hardware might even make it a net loss overall for PRT. Future PRTs might include them as a minor optimization.

    Regenerative braking is much more promising in trains where stops are frequent and the amount of energy recovered per stop is very large.

  3. Hi Transportation Enthusiast

    Thank you for your detailed calcuation. The conclusion is similar. The average occupancy rate is 25% in both cases (instead of 50%).

    The vehicle load per passenger is better for MRT compared to PRT, but after adjusting for distance and stopping, the result may come out to be about the same (but this is based on factors used in my assumptions).

    The benefits of PRT are:
    > short waiting time
    > short journey time
    > seated (instead of standing)


Note: Only a member of this blog may post a comment.