Demand Model Process
In addition to these standard steps, the model also estimates air emissions.
The following description of each modeling step is extracted from, A Transportation Modeling Primer by Edward A. Beimborn at the Center for UrbanTransportation Studies, University of Wisconsin-Milwaukee:
Trip Generation:
The first step in travel forecasting is trip generation. In this step information from land use, population and economic forecasts are used to estimate how many person trips will be made to and from each zone. This is done separately by trip purpose. Trip purposes that can be used include: home based work trips (work trips that begin or end at home), home based shopping trips, home based other trips, school trips, non-home based trips (trips that neither begin nor end at home), truck trips and taxi trips. Trip generation uses trip rates that are averages for large segment of the study area. Trip productions are based on household characteristics such as the number of people in the household and the number of vehicles available. For example, a household with four people and two vehicles may be assumed to produce 3.00 work trips per day. Trips per household are then expanded to trips per zone. Trip attractions are typically based on the level of employment in a zone. For example a zone could be assumed to attract 1.32 home based work trips for every person employed in that zone. Trip generation is used to calculate person trips. These are later adjusted in the mode split/auto occupancy step to determine vehicle trips.
top of page
Distribution:
Trip generation only finds the number of trips that begin or end at a particular zone. These trip ends are linked together to form an origin-destination pattern of trips through the process of trip distribution. Trip distribution is used to represent the process of destination choice, i.e. "I need to go shopping but where should I go to meet my shopping needs?". Trip distribution leads to a large increase in the amount of data which needs to be dealt with. Origin-destination tables are very large. For example a 1200 zone study area would have a 1,440,000 possible trip combinations in its O-D table. Separate tables are also done for each trip purpose.
The most commonly used procedure for trip distribution is the 'gravity model'. The gravity model takes the trips produced at one zone and distributes to other zones based on the size of the other zones (as measured by their trip attractions) and on the basis of the distance to other zones. A zone with a large number of trip attractions will receive a greater number of distributed trips than one with a small number of trip attractions. Distance to possible destinations is the other factor used in the gravity model. The number of trips to a given destination decreases with the distance to the destination (it is inversely proportional). The distance effect is found through a calibration process which tries to lead to a distribution of trips from the model similar to that found from field data.
'Distance' can be measured several ways. The simplest way this is done is to use auto travel times between zones as the measurement of distance. Other ways might be to use a combination of auto travel time and cost as the measurement of distance. Still another way is to use a combination of transit and auto times and costs (composite cost). This method involves using multiplying auto travel times and costs by a percentage and transit time/cost another percentage to get a composite time and cost of both modes. Because of calculation procedures, the model must be iterated a number of times in order to balance the trip numbers to match the trip productions and attractions found in trip generation.
top of page
Mode Choice:
Mode choice is one of the most critical parts of the travel demand modeling process. It is the step where trips between a given origin and destination are split into trips using transit, trips by car pool or as automobile passengers and trips by automobile drivers. Calculations are conducted that compare the attractiveness of travel by different modes to determine their relative usage. All proposals to improve public transit or to change the ease of using the automobile are passed through the mode split/auto occupancy process as part of their assessment and evaluation. It is important to understand what factors are used and how the process is conducted in order to plan, design and implement new systems of transportation.
The most commonly used process for mode split is to use the 'Logit' model. This involves a comparison of the "disutility" of travel between two points for the different modes that are available. Disutility is a term used to represent a combination of the travel time, cost and convenience of a mode between an origin and a destination. It is found by placing multipliers (weights) on these factors and adding them together. Travel time is divided into two components: in-vehicle time to represent the time when a traveler is actually in a vehicle and out-of-vehicle time which includes time spent traveling which occurs outside of the vehicle (time to walk to and from transit stops or parking places, waiting time, transfer time). Out-of-vehicle time is used to represent "convenience" and is typically multiplied by a factor of 2.0 to 7.0 to give it greater importance in the calculations. This is because travelers do not like to wait or walk long distances to their destinations. The size of the multiplier will be different depending upon the purpose of the trip. This is because it has been found that people tend to be more willing to wait or walk longer distances for work trips than for shopping trips.
Travel cost is multiplied by a factor to represent the value that travelers place on time savings for a particular trip purpose. For transit trips, the cost of the trip is given as the average transit fare for that trip while for auto trips cost is found by adding the parking cost to the length of the trip as multiplied by a cost per mile. Auto cost is based on a "perceived" cost per mile (on the order of 5-7 cents/mile) which only includes fuel and oil costs and does not include ownership, insurance, maintenance and other fixed costs (total costs of automobile travel are 25-40 cents per mile). Travelers have been found to only consider the costs that vary with an individual trip rather than all costs when making mode choice decisions.
Disutility calculations may also contain a "mode bias factor" which is used to represent other characteristics or travel modes which may influence the choice of mode (such as a difference in privacy and comfort between transit and automobiles). The mode bias factor is used as a constant in the analysis and is found by attempt to fit the model to actual travel behavior data. Generally, the disutility equations do not recognize differences within travel modes. For example, a bus system and a rail system with the same time and cost characteristics will have the same disutility values. There are no special factors that allow for the difference in attractiveness of alternative technologies.Once disutilities are known for the various mode choices between an origin and a destination, the trips are split among various modes based on the relative differences between disutilities. The logit equation is used in this step. A large advantage in disutility will mean a high percentage for that mode. Mode splits are calculated to match splits found from actual traveler data. Sometimes a fixed percentage is used for the minimum transit use (percent captive users) to represent travelers who have no automobile available or are unable to use an automobile for their trip.
Automobile trips must be converted from person trips to vehicle trips with an auto occupancy model. Mode split and auto occupancy analysis can be two separate steps or can be combined into a single step, depending on how a forecasting process is set up. In the simplest application a highway/transit split is made first which is followed by a split of automobile trips into auto driver and auto passenger trips. More complex analysis splits trips into multiple categories (single occupant auto, two person car pool, 3-5 person car pool, van pool, local bus, express bus, etc.). Auto occupancy analysis is often a highly simplified process which uses fixed auto occupancy rates for a given trip purpose or for given household size and auto ownership categories. This means that the forecasts of car pooling are insensitive to changes in the cost of travel, the cost of parking, the presence of special programs to promote car pooling such as may occur as a result of the clean air act.
top of page
Assignment:
Once trips have been split into highway and transit trips, the specific path that they use to travel from their origin to their destination must be found. These trips are then assigned to that path in the step called traffic assignment. Traffic assignment is the most time consuming and data intensive step in the process and is done differently for highway trips and transit trips. The process first involves the calculation of the shortest path from each origin to all destinations (usually the minimum time path is used). Trips for each O-D pair are then assigned to the links in the minimum path and the trips are added up for each link. The assigned trip volume is then compared to the capacity of the link to see if it is congested. If a link is congested the speed on the link needs to be reduced to result in a longer travel time on that link. Changes in travel times means that the shortest path may change. Hence the whole process is repeated several times (iterated) until there is an equilibrium between travel demand and travel supply. Trips on congested links will be shifted to uncongested links until this equilibrium, condition occurs. Traffic assignment is the most complex calculation in the travel modeling sequence and there are a variety of ways in which it is done to keep computer time to a minimum.
Considerations of time of day are also important. Traffic assignment is typically done for peak hour travel while forecasts of trips are done on a daily basis. A ratio of peak hour travel to daily travel is needed to convert daily trips to peak hour travel (for example it may be assumed that ten percent of travel occurs in the peak hour). Numbers used for this step are very important in that a small change in the values assumed will make a considerable difference in the level of congestion forecasted on a network. Normally the modeling process does not deal with how traffic congestion dissipates over time.
top of page
 |
 |
 |
 |