Demand and Trips
Demand
Demand consists of
- Demand Divisions
- Profiles
- Matrices (Directed) or Origin Volumes & Splits (Undirected)
Demand Division
A Demand Division defines a group of Types and assigns a proportion to each Type, with the proportions summing to 100%. The demand division is then associated with a demand matrix or origin volume.
Profiles
A Profile can be used to vary the rate at which demand is released over time. A Profile has a Term, defining start time and end time, a number of intervals, and a weight for each interval, specified as a percentage. The weights for all intervals must sum to 100%.
Directed Demand (OD Matrices)
Directed demand is a set of trips, where each trip has an origin and a destination. This demand is entered into the model as a two-dimensional table known as an Origin-Destination matrix (OD matrix). In Traffic Analyst a matrix can be for people, vehicles or freight. A person matrix allows each person to use all modes of transport available to them, a vehicle matrix is used to generate vehicle trips. Vehicle OD matrices can be used to generate background traffic where no mode switch is required.
Freight is modeled as an agent in Traffic Analyst with no walk speed, The freight has route choice and moves along moving walkways and on available modes of transport.
Traffic Analyst has a two-level demand that allows you to enter as many person matrices, freight matrices and vehicle matrices as required.
Undirected Demand (Origin Volumes and Splits)
Undirected demand is made up of two elements: a one-dimensional origin volume and a set of splits. The origin volume specifies the number of people or vehicles to be released from each area or zone, respectively. The splits are used for route assignment at each route choice location – for example at intersections or walkway junctions. Undirected demand can be used for smaller networks to speed up the demand building process. When using undirected demand, it is important to recognize that each person or vehicle does not have a destination while traveling through the network; it discovers its destination only when it arrives.
Each route choice location automatically creates a “Split” object, that by default assigns an equal proportion of all incoming traffic to each of the available exit options. At a 4-way intersection, an incoming agent has a choice of left, right or straight ahead, and the default split object will stochastically assign the agent to one of those exits, each with equal probability of 1/3. This means that if there is a loop in the network, an agent may be directed around the loop, arriving back at the same point. With typical splits, the probability of this happening is low, and in some networks, following a loop around may reflect reality, for example where a pedestrian is window shopping. Some worked examples on the following pages illustrate the probabilities of following a loop.
The Split function is very simple; if more complex routing rules are required, then the route choice tool allows you to create rules that assign the exit (left, ahead, right) based on origin area, agent type, sign settings, etc. The outcome of a route choice rule can be a single exit or a stochastic distribution across multiple exits, similar to a Split.
Trips
Within Traffic Analyst, the Trips component contains a specification of all the trips that will use, or try to use, the capacity provided by the network.
There are several tables of trips within the Trips component:
- Undirected vehicle-trips (Origin only)
- Directed vehicle-trips (Origin-Destination)
The Trip components are used to generate Trips to simulate on the model network.
Every trip contains the following fields:
- A unique name
- A departure time
- A vehicle type
- Origin and/or Destination
- A DNA string defining the decision-making behavior of the agent making this trip