9  Introduction

SPC is divided into two phases. The data preparation phase relies on scripts that only need to be run once. It outputs a postprocessed version of all the raw data sources that allows the model to run smoothly on custom areas. The second phase involves the user choosing a custom area and launching a simulation. It pulls the relevant datasets among the data prepared by the first phase, calculates the different daily activities and formats the results into a single protocol buffer file.

In this section, you can find:

• A step by step description of each element of the SPC pipeline
• The concepts supporting the modelling methods
• A description of the schema of the protocol buffer
• A list of the data sources used to create each data field

Note that due to the large impact each section has on the choices made in other sections, it may be necessary to frequently segue between sections to get a precise understanding of the model.