Big changes are taking place in the way energy companies supply gas to their customers.

With many of the world's easily-accessible gasfields already developed, producers are being forced to look at increasingly remote prospects.

Making these prospects economic is a daunting challenge, particularly when some are hundreds of miles offshore and with no market for the gas close to hand.

One idea gaining credence is to super-cool gas at its point of production and then ship it directly to its end-user.

While so-called liquefied natural gas (LNG) projects are now well established onshore, no floating LNG (FNLG) terminal has been achieved to date.

Royal Dutch/Shell, the Anglo-Dutch oil company, is leading the way in developing the world's first floating LNG scheme. It has put forward a proposal for its Greater Sunrise field in the Timor Sea that, it claims, could lower the cost of development by as much as 40 per cent.

Marvin Odum, director of Shell Gas and Power, believes FLNG schemes could revolutionise the gas market.

"FLNG will become part of a tool-kit available to energy producers for monetising stranded gas and I expect we will see many of these schemes being developed in the future," he says.

Existing LNG schemes use conventional pipelines to carry gas from offshore production platforms to a liquefaction plant onshore. Here the gas is cooled to minus 160 degrees celcius before being transferred as a liquid to specialised tankers and shipped to the customer, where it is re-gasified.

The problem is that some fields are too remote to economically build a pipeline. Shell's Greater Sunrise scheme would require a 500km pipe from the Bonaparte Basin of the Timor Sea to Darwin in Australia, pushing the cost of the project to about $4.9bn.

Mr Odum says FLNG could help reduce the environmental impact of extracting gas, as well as lowering costs. It would minimise the need to disturb the seabed and would reduce the amount of carbon dioxide emissions produced by eliminating the use of large jet engines normally needed to drive the gas to shore.

"In some cases it may be preferable to construct FLNG projects even for fields relatively close to land. Equally, a number of people are now looking at the possibility of floating re-gasification plants at the customer end of the business, especially in the US where it is difficult to get planning permission for onshore plants," he says.

The individual components of an FLNG scheme are similar to the well-established technologies used in LNG tankers and onshore LNG plants for years. But pulling them all together is proving to be a complex task for engineers.

Shell's Greater Sunrise proposal envisages a vessel 400m long by 70m wide, making it larger than a US aircraft carrier. This sounds big but is a fraction of the size of onshore LNG plants, which normally cover an area of half a kilometre square.

Safety is a prime concern as people will be living on-board to operate the vessel. Sophisticated computer modelling has been used to work out the safest lay-out of pipework and plant, minimising the possibility of a serious explosion. In addition, Shell's research and development division has carried out physical modelling of the sea conditions the vessel would have to endure to test the docking and loading system for LNG tankers.

The liquefaction process for the FLNG has also been adapted from that of a traditional LNG plant. Conventional propane pre-cooled mixed refrigerants have been ruled out due to their large hydrocarbon contents, which would be unsafe on a confined vessel.

Instead, Shell has developed a special mixed refrigerant liquefaction process that will reduce hydrocarbon content and simplify the overall design and operation of the plant.

For LNG production rates of up to 2m tonnes a year, a single mixed refrigerant process will be used, but for larger through-puts of up to 4m tonnes a year, Shell has a proprietary dual mixed refrigerant process.

Cambridge Energy Research Associates, an energy consultancy, said in a recent global LNG study that floating schemes would be in the frontline of technologies to drive the cost of gas production down.

Shell still has to convince Phillips Petroleum - its project partners on the Greater Sunrise field - of this. However, if all goes to plan, the first FLNG vessel could be in production in 2007.

The group is also considering an FLNG vessel for the large Kudu gas field off the southern Namibian coast, as well at a number of other locations that it is keeping secret for the time being.