Demand generation

Trip generation

The demand generation module generates the demand for the transportation simulation system. Two important methods are: (i) origin-destination matrices, and (ii) activity-based demand modeling.

Origin-destination (OD) matrices are the more traditional method. OD matrices contain the number of trips from $n$ starting points to $n$ destinations; it is therefore an $n \times n$ matrix. These matrices can refer to arbitrary time periods. Until a couple of years ago, one typically used 24-hour time periods; these days, people often concentrate on ``morning peak'' and ``afternoon peak'' periods since the main direction of travel is obviously different between these periods.

In many situations, it is desirable to have information about demand generation that goes beyond OD matrices. In such situations, the more far-reaching method of activities-based demand modeling is an alternative. Here, the simulation includes models of human behavior with respect to the planning of a day. This includes where and when to eat, sleep, work, shop, etc. For example, a person may start the day at home, be at work at 8am, work for eight hours, go shopping which takes an hour, then be at home for the rest of the day. Assuming that all the transportation pieces take half an hour, this would fix the transportation schedule to: leave home at 7:30am, be at work at 8am, leave work at 4pm, arrive at shopping at 4:30pm, leave shopping at 5:30pm, arrive home at 6pm.

Once the simulation ``knows'' where and when people do their activities, transportation is generated via connecting activities that take place at different locations. Note that it is not necessary (and probably not possible) to forecast such activities for specific persons; however, there is hope that we will be able to get useful ensemble averages similarly to Statistical Physics.

Figure 2.2: Illustration of a daily activity plan.

Route generation

Once trips (e.g. starting times, starting locations, and destination locations) are known, the exact transportation for these needs to be generated. This includes mode choice (walking, bicycle, train, car, etc.) and the precise routing. The output of this module are complete plans for each individual in the simulation.

Figure 2.3: Illustration of a daily plan including routes.