Common Limitations and Issues

The following are some of the common limitations that might be of concern in the generation of trips.

Decisions are independent: Travel behavior is a complex process in which one household member’s decisions are often contingent on others within the household. Child-care needs, for example, may influence how and when people travel to work. This interdependence for the making of trips is not taken into account.

Trip purposes are limited: A simpler trip pattern results in no more than four to eight trip finishes. All shopping trips are treated the same whether grocery or lumber shopping is done. Flome-based “other” trip purposes cover a wide range of purposes (e.g., medical, visiting friends, banking, etc.) that are affected by a broader range of factors than those used in the modeling process. Limited variables: Trip making is seen as a function of just a few variables including auto ownership, household size, and employment. Generally, other considerations such as the nature of the transit service, ease of walking or cycling, fuel prices, land use design, and so on are not included. Combinations of trips (trip chaining) are ignored: Travelers can often combine a variety of purposes into a sequence of trips as the errands run and connect activities together. This is called the chaining of journeys, and is a complex process. The modeling process treats such combinations of trips in a very limited manner. Non-home-based journeys, for example, are calculated based only on work characteristics of zones and do not consider how members of a household manage their errands.

Feedback, cause-and-effect problems: Often trip generation models measure trips as a function of variables that might in effect depend on how many trips there are. For instance, shopping trip attractions are seen as a feature of retail jobs, but it could also be argued that the number of retail employees in a shopping center may depend on how many people come to shop there. The question of “chicken and egg” often arises in travel predictions, and is difficult to avoid. Another example is that traveling depends on auto availability, but it could also be argued that the number of cars owned by a household will depend on how' involved they are when making travels. Trip generation only finds the number of trips that begin or end at a particular zone. Through the cycle of trip distribution these trip ends are joined together to create an origin- destination pattern for trips. 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?”). Distribution of trips leads to a substantial increase in the amount of data that needs to be addressed.

Quite wide origin-destination tables: For example in its O-D table a 1,200-zone study area would have a potential 1,440,000 trip combinations. For each intent of the trip, separate tables are also done. The gravity model is the most widely used trip, distribution method. The gravity model takes the journeys produced in one zone and distributes them to other zones based on the size of the other zones (as measured by their travel attractions) and the distance to other areas. A zone with a large number of travel attractions would earn more dispersed travels than one with a small number of travel attractions. The other element used in the gravity model is distance from the possible destinations. The number of trips to a given destination decreases with the distance (it is inversely proportional) to the destination. The distance effect is discovered through a calibration method that seeks to lead to a similar distribution of trips from the model, as observed from field data.

Transportation models are called upon to provide predictions for a diverse set of issues that may go beyond their capabilities and original purpose in some cases. Travel demand management, trip reduction programs focused on the workplace, pedestrian and cycling systems, and land use policies may not be treated well in the process. Models of transport travel forecasting use packaged computer programs that have limitations on how easily they can be changed. In some cases, the models can be updated to fit additional factors or procedures (quick fix) while significant changes are required in other cases or a new software is needed. The following are some possible model modifications that may help to improve their usefulness:

All models are based on travel patterns and conduct data: If that data is obsolete, the findings will be incomplete or incorrect, no matter how good the models are. One of the most important w'ays to improve the quality and reliability of the model is to have a good basis for using recent data to calibrate the models and to ensure that their accuracy is tested. Models need to prove that they provide an accurate picture of current travel before using them to predict future travel. An improved representation of bicycle and pedestrian travel is needed. In modern travel, demand models commuting by bicycle and walking is not done well. Better approaches are needed to deal with these types of journeys. This can be done by incorporating factors in models of trip generation that relate the making of trips to the amenities of the pedestrian or bicycle. These types of trips could also be expanded to include mode choice methods.

Better auto occupancy models: Current practices for auto occupancy appear to be insensitive to a wide range of policies which may contribute to more or less carpooling. Auto-occupancy policies need to be responsive to the cost of parking and travel costs as well as the number of trips that occur between a destination and origin. Also, it may be desirable to treat ride sharing among family members differently than carpooling between persons from different households. Procedures that increase the number of trip purposes to deal with segments of the market that are likely to share trips may help with this issue.

Better time of day factors: Congestion rates in hours other than the peak period are required to gain a better understanding of the essence of congestion as it persists throughout the day and into the future over time. Hourly conversion factors must be scrutinized very carefully to ensure that they represent actual traffic variations.

Use more trip purposes: Additional travel purposes (market segments) may provide a way to better represent complex household travel patterns and chaining trips. That would also provide procedures for the generation of trips that are sensitive to more factors that would result from travel management techniques.

Better representation of access: The transport models are not well reflected in land use policies that promote transit use or provide high-quality site design with good pedestrian access. Improved methods are needed to measure the disutility of the transit and highway travel access portions. These approaches w'ould include calculating a sensitive access index in areas that used more bus/pedestrian/bike friendly architecture for easy access and waiting for transit vehicles.

Incorporate costs into trip distribution: Trip distribution models should use a generic distance calculation that involves travel costs by various means including cost of parking. Therefore, these models will help show the sensitivity of travel patterns to cost changes.

Add land use feedback: It is important to take measures to close the forecasting process loop in order to enable a better representation of the land use and travel demand interactions. Simulation of land use models should be applied to the model series to help determine how a new transport system can lead to changes in land use.

Add intersection delays: Most congestion is experienced in an urban traffic network at traffic signals or stop signs rather than on intersection roads. Travel forecasting models should include routines that calculate the intersection delay encountered. Furthermore, splits of the intersection signal should be viewed as a variable that would be changed as the cycle of traffic assignment iterates to reach equilibrium.

Vehicle Routing Problem

A dispatcher managing a vehicle fleet is often required to make the vehicle routing decisions. Another such decision is how best to assign a group of customers to a vehicle fleet, and how to arrange and plan their visits. The aim of solving these Vehicle Routing Problems (VRP) is to provide a high level of customer service by accommodating any time frames while keeping the total operational and investment costs as low as possible for each route. The limitations are to complete the routes with the resources available and within the time limits placed by driver work shifts, driving speeds, and customer commitments. Network analysis offers a problem- solver for vehicle routing, which can be used to find approaches for such complex fleet management tasks. Consider an example of delivering goods from a central warehouse location to the grocery stores. There is a three-truck fleet at the factory. The warehouse works only within a given window' of time from 8:00 a.m. to around 5:00 p.m. w'hen all trucks have to return to their warehouse. That truck has a 15,000- pound capacity, which limits the amount of goods it can carry. Every store has a requirement for a certain quantity of goods (in pounds) that must be shipped, and each store has time periods that restrict w'hen deliveries are due. In addition, the driver can work just 8 hours a day, needs a lunch break, and is charged for the amount spent on driving and servicing the shops. The purpose is to create an itinerary for each driver (or path) so that deliveries can be made w'hile complying with all the service requirements and reducing the driver’s total time spent on a particular route. The following figure shows three routes obtained by solving the aforementioned vehicle routing problem.

5.5.2.1 Location-Allocation

Location-allocation helps you select which facilities to operate from a set of facilities based on their potential interaction with the demand points. It can help you answer such questions as:

In view of a set of existing fire stations which site would provide the best response times for the community for a new fire station?

If a retail company has to scale down, which stores should it close to retain the most competition overall?

Where will a plant be designed to reduce the gap to distribution centers?

Facilities would represent the fire stations, retail stores, and factories in these examples; demand points would represent buildings, customers, and distribution centers.

5.5.2.2 Time-Dependent Analysis

All of the solvers allow you to integrate live and historical traffic data into an analysis so that you can find the best route for a given time of day; determine the best place to plan an ambulance at 8:00 a.m., 12:00 p.m., 4:00 p.m., etc.; and create service areas for different times of day. With various dates and times, the outcomes of any study may vary, because traffic conditions and travel times can change.

Concluding Remarks

Micro-level investigations and real-time analysis proves that demography affects crime, catering to the various parameters of intervention such as impact of age, sex, occupational structure, and society on criminal behavior, and the difference between compositional and contextual effects of demographic structure on aggregate crime rates. The intersection of criminal and demographic events in the life course; the influence of criminal victimization and aggregate crime rates on residential mobility, migration, and population redistribution are some pertinent issues of analysis. Geographical concerns and events have both a spatial and temporal dimension highlighting the approaches of network analysis used in criminal intelligence to understand and act against serious crime occurrences, criminal groups, and criminal markets. Transportation models, travel demand management, trip reduction programs, and models of transport travel forecasting are also pertinent features of geospatial analysis in crime investigations.

References

Conrad, T. and Irene. В. T. (1958) The Changing Population of the United States. John Wiley and Sons. New York. London. 12-16.

Donald, J. B. (1969) Principles of Demography. Wiley, 32.

Gordon, A. C. (2016) Fundamentals of demographic analysis: Concepts, measures and methods. Springer Series on Demographic Methods and Analysis, 38(2).

Herve, L. B. (2008) The Nature of Demography. Princeton University Press, Social Science. 361.

Pathak. L. (1998) Population Studies: The Discipline, Development Pattern and Information System. Rawat Publications. 17-18.

Scott. J. S. and Steven. F. M. (2000) Crime and Demography: Multiple Linkages. Reciprocal Relations. Ann. Rev. Sociol. 26, 83-106.

Scott. J. S.. Katherine. T., and Sunita. B. (2014) Skewed sex ratios and criminal victimization in India. Jastor, 51(3), 1019-1040.

Guttentag and Secord. (1983) Too Many Women. Sage Publications. 277.