PROMISE: Policy Sensitive Rail Passenger Forecasting for the Netherlands Railways
COHN N and DALY A and ROHR C, Hague Consulting Group and DAM A, OOSTERWIJK W and VAN DER STAR T, Netherlands Railways, The Netherlands
The Netherlands Railways (NS) currently uses the Netherlands National Model System (LMS, Rijkswaterstaat, 1990) for forecasts and evaluation of important projects. The LMS is the Netherlands government's transport forecasting model system, and is used to
The Netherlands Railways (NS) currently uses the Netherlands National Model System (LMS, Rijkswaterstaat, 1990) for forecasts and evaluation of important projects. The LMS is the Netherlands government's transport forecasting model system, and is used to determine the effects of government policy and exogenous developments on the mobility of the Netherlands population. The LMS carries with it a number of disadvantages which are of importance to the NS. In the ftrst place, the LMS is not able to provide information about train travel at the level of detail needed by the NS.
Second, the model was developed primarily for analysis of travel by car and focuses on that mode. Third, it appears that complete dependence on an external organisation (Hague Consulting Group developed the LMS for the Netherlands government and continues to operate the model) for making forecasts can sometimes be a hindrance.
The new organisational structure of the NS has led to a need for producing new forms of information. This specific information must serve as input for better design of the NS product to meet clients' wishes in more detail than previously. It is necessary that information be available at the train route level as this is to an extent the level at which management occurs. NS policy is also focused on new markets. In the future an important issue will be the provision of a complete travel chain, whereby the NS provides a door-to-door travel product.
The following conclusion is clear: the NS requires an in-house forecasting model for train travel, a model which provides tailored information, produces forecasts for the medium term (5 to 10 years), and is easy to connect to existing NS models.
The model is intended to be an important tool for:
* determining the effects on train use of changes in train fares;
* determining the effects on train use of developments in the train service schedule;
* exploring the effects of macro-economic developments; and
* drafting marketing plans for the medium term. The name of this model system is ProMiSe. With this model we hope to forecast the 'promise' of train travel in the Netherlands. ProMiSe is an acronym for
"Prognosemodel voor Middellange termijn Studies" forecasting model for medium- term studies).
Association for European Transport