The Ariane 4 launch vehicle has recently passed milestone in it's history. The rocket, designed in the mid seventies, launched it's 100th payload. It will soon be retired, to be replaced by the Ariane 5. The reason for the retirement, growth in the size of payloads. Actually when the Ariane 5 was proposed in the late eighties, it was considered to be to big for the market, but now there is an upgrade program to increase it's payload capacity even as it begins it's career. This points out one of the basic facts of launch vehicle design, any design will soon be obsolete due to rapid changes in the marketplace.
The big push for a Reusable Launch Vehicle is based on the economics of a high flight rate, with 100 flights a year as being the common figure quoted for break- even, and 200 needed for profitability. The current plans for a RLV also call for it to be an unmanned vehicle to place payloads into Low Earth Orbit. There are two obvious problems here. One, the Ariane 4 has been dominate in the market, yet it took twenty years for it to achieve 100 flights. That's 5 flights a year. There is no market available for a 200 flight per year RLV. Two, any design for a RLV used to transport cargo will almost immediately become obsolete due to payload growth.
Space is a marketplace in flux, it is changing rapidly, (despite NASA's glacial pace of change), and the payload requirements are extremely unpredictable. If you have no idea what size and weight of payload you will be kicking into orbit in two years, you need a launch vehicle system that is extremely flexible. The Ariane 4 is actually a group of vehicles, with several different versions consisting of different upper stages and boosters. The Ariane 4's flexibility has been the reason for it's being able to hang on for twenty years. What is needed for cargo launches in a rapidly changing market place is a program that can produce an entirely new vehicle, fitting the customers specifications, quickly and at very low cost. This is a situation crying out for a Minimum Cost Design launch vehicle. With MCD, you would redesign your launch vehicle every two years to precisely fit the marketplace. If you missed the market, you simple redesign the vehicle again. The idea is to keep the cost of changes down.
This brings us back to the current big push for a RLV. A RLV has fixed payload size, no flexibility, no growth capability , very high investment, limited fleet expansion capability, and must fly nearly full to make costs. Obviously a RLV is NOT the vehicle you want for space cargo flights. What a RLV needs is the one payload that hasn't changed in 100,000 years, the one payload that will appreciate the technology of an RLV, the one payload that will pay the extra price for an RLV, humans.
So, how many humans do we need in orbit? I'll be really generous and optimistic here and say that we will need 200 people a year into orbit. Divided by the flight rate of a RLV (100), that means a vehicle capable of carrying two people on each flight. This points to a very small vehicle, much smaller than the current proposals, and a "fleet" requirement of only two vehicles at most. If you're a big RLV supporter, this might seem a dismal prospect, but there are also good points to this size vehicle. One, a two passenger vehicle is very small, and very cheap, and very easy to develop. Two, most of the X-prize vehicles fit exactly into this size range, so there lots of proposals. The one big problem is getting NASA out of the launch vehicle business and letting private enterprise develop this smaller market friendly RLV.