Car-sharing as an alternative to single-occupancy driving is quite common, and is promoted by many companies (like BA and BAA) to reduce their carbon footprint.

It does rely on people making much the same trip at much the same time on most days.

I saw a study of a system which could be used on a much more off-the-cuff basis. It was in Transportation Research Record 2217, in an article entitled "Real time ridesharing" by Andrew Amey and colleagues.

The aims of the project were to identify, highlight and discuss potential benefits and obstacles: it found a series of challenges.

The study looked at the idea of one-off ride-shares, arranged at relatively short notice. Clearly, these have become much easier to arrange with developments in technology (in particular, smart phones with constant network connectivity and Global Positioning Systems).

Research among stakeholders (including commercial ride-share service providers and public-sector promoters) was followed by a workshop session.

Economic challenges include information (about each other - driving record, criminal history), transaction costs (the time needed to set up the share, and the need for a diversion away from a preferred route), and the decreasing cost of car purchase (at an all-time low of 22 weeks income, compared with highs of 30 weeks 10-15 years ago).

Social and behavioural challenges include 'stranger danger' (but most car-shares are between people known to each other), power mismatch (the driver is 'in charge' and to a large degree able to dictate things like stops, routes and timings), flexibility and reliability.

Institutional challenges include institutional collaboration (difficult!), the business model (from the point of view of a commercial coordinator) and competition.

The study found that relatively few people wanted to arrange a ride at short notice: the need was more for a part-time or occasional arrangement made perhaps the evening before. This suggests that a real-time service could close at, say, 20:00 and send requests and offers out then.

Large employers are obvious potential organisers and beneficiaries.

An interesting idea, on which more work is necessary.