After reading Ramyata Limbu's "No go on green" (#104) I want to draw your readers' attention to the fact that the primary objective of providing traffic signals in "unsignalised" intersections is to reduce accidents by reducing conflict among different vehicular movements.

The last time I saw a properly signalised intersection in Kathmandu was at Thapathali. There was no power supply and a traffic personnel was directing the traffic. I have since been reading reports of the recent signalisation of intersections in Kathmandu, and how most residents hate it because of slower traffic movement.


The effectiveness of traffic signals in reducing traffic related accidents at intersections is in no doubt. However, in downtown locations where traffic signals are closely spaced, and in the absence of synchronisation, signals can cause congestion, slow vehicular movement and increase mid block accidents between intersections. But, they will still reduce accidents at the intersections. The agencies involved in implementation of signalisation must have taken into consideration these facts and also must have weighed the pros and cons.

Limbu also points out that a simulation could have been performed before installing the actual hardware. This is what is done in practice and is called microscopic traffic simulation. But computer simulation requires huge quantity of data that is able to describe existing traffic conditions (volume, turning movements, speed etc) in "microscopic" detail. I don't think any agency in Nepal has that kind of data.

Most simulation software produced in developed countries is designed for entirely different traffic conditions in terms of composition of traffic volume, pedestrian and driver behaviour. Even if such sophisticated software is used to simulate traffic conditions of roads in Nepal, it will not be able to generate statistically correct results. We don't know what kind of simulation or detailed studies were done before designing the traffic signal system for Kathmandu, but some kind of simulation exercise must have been carried out.

Having said this, traffic accidents at intersections in Kathmandu are certainly going to decrease, and with synchronisation traffic movement will also improve. But this will also depend on the performance of the overall system including the response of drivers to the system.

Rajat Rajbhandari, New Jersey Institute of Technology
Interdisciplinary Program in Transportation