Specifying the appropriate levels of demand is key to the legitimacy of a LEGION Model. Demand refers not only to the number of pedestrians in, entering or leaving a site, but also to the type of pedestrians visiting the site, for example: commuters, tourists, etc. The composition of the population of a site is defined using Entity and Supply Types (see About Entity and Supply Types ).

The quantitative component of demand is typically presented in the form of an Origin-Destination matrix and associated Arrival Profiles. The combination of data describes:

• How many pedestrians occupy the site at the start of the simulation period (if any).
• How many pedestrians enter the site during the period to be modelled.
• When pedestrians enter – this can be aggregated over the entire model period or broken down into specific intervals.
• The entry points used by the pedestrians.
• The proportion of pedestrians going to various destinations within the site.

Demand Data can be defined and introduced into a LEGION model in one of two ways:

• The absolute number of pedestrians on site at the beginning of the modelled period. This is introduced by using Populated Zones and Evacuation Zones (the number of pedestrians can be specified in a Population Profile or entered manually when adding a zone. See Data profiles worksheet for more information).
• The arrival of pedestrians over time, distributed over the duration of the modelled period. This is introduced using Entrances and associated Arrival Profiles.

Demand Data may not always be complete or available in the correct format. You may need to manipulate or extrapolate the data to create as complete a data set as possible.

It is advisable to test multiple scenarios during modelling to ensure that the space is tested at the extremes of its demand limits (see Scenarios for further information). In this way, you can interpret results in the knowledge that the space is being pushed to a level where any identified problems (congestion, blockages) can be assumed to be the 'worst case'.