This is a small-capacity solution for low-frequented routes, in particular in rural areas; thus a replacement for a bus or a small branchline DMU (their webpage focuses on the latter comparison). It is not a replacement for high-capacity vehicles on busies routes and certainly doesn't eliminate the need for track construction, even if maximum axleload is just 5 t (they only claim that there is no need to dig below the road base).

I was merely thinking that there is a role for simple solutions which can be rapidly implemented without vast cost and complexity.

This can be implemented without vast cost and sensibly only on branchlines without passenger trains (or in threat of losing them) with track in relatively good shape. Due to its limited capacity, people mover sized ULR can only be complementary to normal trams, too (not just heavy rail). For such uses in cities and for uses on new lines in rural areas, the real question is then the (new) track, the construction of which will be neither cheap (compared to the cost of the vehicles) nor simple. While it will likely be cheaper than normal heavy or light rail, the ridership and ticket income will be lower, too. So the costs need to reduce to scale, and I am very sceptical about that, they need to prove it in practice.

_{*Lunatic*, n.
One whose delusions are out of fashion.}

This shouldn't be surprising. The internal combustion engine has losses a gas turbine doesn't: air compression, piston movements. In addition, a diesel engine on a public transport vehicle is constantly revved up and down, while a power plant gas turbine is operated closer to the ideal range, thus practical average thermal efficiency differs from the maximum more. Electricity transmission losses are only a few percent.

Here is one source with a concise summary:

Trolleybus UK

The thermal efficiency (electricity out/fuel in) of modern combined cycle thermal power stations is essentially constant at around 60% (i.e. 40% gets lost to heat, not the 70% mentioned elsewhere), very large (marine) diesels do almost as well at around 55%, automotive diesels (bus engines) do about 40% at best but less at part load. In a diesel bus operating an urban duty cycle and unlike a trolleybus there are significant losses in fluid element automatic transmissions, while idling and moving at low speeds and operating at well below engine maximum efficiency levels, while burning fuel to regenerate particulate traps, etc. The efficiency of a diesel bus at the road wheels is a lot less than engine peak efficiency might suggest.

...Including all the grid, substation, overhead wiring, on-vehicle etc, losses, and regeneration gains for a trolley and all the transmission, cooling, idling and part load running losses, etc, for diesel, a diesel bus is somewhere around 25% efficient at the wheels based on typical MPG and a trolley around 40%.

For the "around 60%" figure for (combined-cycle) gas plants, here is the state-of-the-art in 2010:

CCGT: Breaking the 60 per cent efficiency barrier - Power Engineering International

GE, of course, announced the achievement of that lofty goal several years ago. But that turned out to be a little premature. These days, GE is a lot more conservative - by far the most conservative of the major OEMs - at least with regard to efficiency announcements.

Siemens and Mitsubishi Heavy Industries Limited (MHI), on the other hand, are claiming an imminent victory. According to Carlos Koeneke, technical director at Mitsubishi Power Systems Americas, the company's J-class machine will provide in excess of 61 per cent by 2011. In the meantime, the Japanese company leads the pack with 59.1 per cent verified on an M701G2 gas turbine at the 1500 MW Tokyo Electric Kawasaki power station in Japan.

Siemens claims that it can match this figure. Fischer states the Siemens F class in combined cycle operation scores in the 58 per cent to 58.7 per cent range at ISO conditions 1-on-1 configuration. Recently they reached well over 59 per cent in Irsching 5 which is a combined-cycle gas turbine (CCGT) power plant in a 2-on-1 configuration based on two SGT5-4000F with a special optimized cycle design. In future, the coming H-class will offer above 60 per cent," says Willibald Fischer, programme manager for the Siemens SGT5-8000H. After undergoing heavy testing at the Irsching prototype plant in Germany the machine will be released on the market. The company is so confident of its numbers that it is virtually guaranteeing 60 per cent for current orders.

Note that the figure can be even higher if the plant supplies distance heating, too, and heating use is included in the efficiency number:

Austria's most efficient thermal power plant goes on line - Hand-over ceremony for the Mellach combined-cycle power plant - Siemens Global Website

In the opening ceremony held on June 22, 2012, Siemens officially handed the Mellach combined-cycle power plant (CCPP) over to the Austrian power provider VERBUND Thermal Power GmbH & Co. KG. The heat-and power cogeneration plant had already commenced operation in May 2012. With an electrical rating of 838 MW and an efficiency of 59.2 percent in straight power generation, the new plant surpasses the contractually agreed performance values. Thanks to the district heating output of 400 MWth, more than 80 percent of the energy in the fuel is put to effective use.

For the "about 40%" figure for bus diesel maximum thermal efficiency, here is a state-of-the-art number:

Scania Delivering 85 Ethanol Buses to Stockholm Suburbs; E95 in a Diesel Engine | OnGreen

The third-generation ethanol engine is an adaptation of Scania's 9-liter diesel engine with charge-cooling and exhaust gas recirculation (EGR). The engine delivers 270 hp (201 kW) of power and torque of 1,200 N·m (885 lb-ft), and offers a thermal efficiency of up to 43%, compared to thermal efficiency of up to 44% for diesel, according to Scania.

For transmission losses, here is an overall number which is a national average for the USA:

How much electricity is lost in transmission and distribution in the United States? - FAQ - U.S. Energy Information Administration (EIA)

According to EIA data, national, annual electricity transmission and distribution losses average about 7% of the electricity that is transmitted in the United States.

You've got a very good point about splitting out industrial CCS from other industrial interventions.  It's a tricky thing, decarbonising industrial processes - for example, what we will do about things like emissions from clinker production for cement. Splitting out the CCS element in the model will help untangle how big an issue that is.

Energy efficiency of industry ... well, I've seen some good stuff in practice - Paris MinesTech and ECLEER do some great work in this field - particularly with heat recovery. And I know of one steel plant that cut 10% of its energy bill just with one improved computer heat flow model. So there is some real potential there.

There's a good economic argument to be made, I think, that industry has inevitably become excessive energy consumers, because the price they directly see for energy has effectively been subsidised, both by government tax breaks and payments to fossil fuel industries, and by the externalities of fossil fuels. And it's reasonable to believe that removing those subsidies would bring about energy-efficiency improvements in industry. But that doesn't tell us how much energy we'd save. I agree with you that a lot of industry forecasts on that question are to be treated with suspicion.

Yeah, synfuels have to be one of the next big things in it.  First, I'll get multi-fuel district heating working; then I'll have a look at syn fuels.

If anyone can point me at some references for efficiencies and costs, I'd really appreciate it.  I'm in discussions with the team for the original DECC model too, so will pass these on upstream too.