Transport Demand Management in Turkey: A Genetic Algorithm Approach

Abstract

This article proposes new models for estimating transport demand using a genetic algorithm (GA) approach. Based on population, gross national product and number of vehicles, four forms of the genetic algorithm transport planning (GATP) model are developed – one exponential and the others taking quadratic forms – and applied to Turkey. The best fit models in terms of minimum total average relative errors in the test period are selected for future estimation. Demand management strategies are proposed based on three scenarios: restricting private car use, restricting truck use and the simultaneous management of private car use and goods movement. Results show that the GATP model may be used to estimate transport demand in terms of passenger-kilometers traveled (pass-km), vehicle-kilometers traveled (veh-km) and ton-kilometers completed (ton-km). Results also show that the third scenario – simultaneous restrictions on private car use and goods movement – could reduce total veh-km by about 35% by 2025 in this study of Turkish rural roads.     

The applicability of non-cooperative game theory in transport analysis

Various models that incorporate concepts from Non-Cooperative Game Theory (NCGT) are described in the transport literature. Game Theory provides powerful tools for analysing transport systems, but these tools have some drawbacks that should be recognised. In the current paper we review games that describe transport problems and discuss them within a uniform context. Although the paper does not introduce new tools, it presents insights concerning the relations between transport models and games. We divide existing games into groups and show that some common features characterise multiple games. We distinguish between games that make a conceptual contribution and games that are suitable for application. Compact or symmetric game structures make remarkable observations but often do not support actual decision-making. Less aesthetic formats, most of which are Stackelberg games between authorities and travellers, are stronger as instruments that assist in determining real-life policies; these formulations can be treated by practitioners as mathematical programs with equilibrium constraints and not as games. Yaron Hollander is currently conducting economic research of bus reliability at the Institute for Transport Studies at the University of Leeds. He previously worked for the Technion—Israel Institute of Technology; for the Israel Institute for Transportation Planning and Research; and for the public transport department at Ayalon Highways Co. Joseph N. Prashker is a professor at the Faculty of Civil and Environmental Engineering at the Technion—Israel Institute of Technology. Till recently he served as head of the Transportation Research Institute at the Technion. His interests are behavioural demand models, network analysis, and Game Theory applications in transportation.     

Deciding whether and how to improve statewide travel demand models based on transportation planning application needs

Many states in the USA have developed statewide travel demand models for transportation planning at the state level and along intercity corridors. Travel demand models at mega-region and provincial levels are also widely used in Europe and Asia. With modern transportation planning applications requiring enhanced model capabilities, many states are considering improving their four-step statewide demand models. This paper synthesizes representative statewide models developed with traditional four-step, advanced four-step, and integrated micro-simulation methods. The focus of this synthesis study is as much on model applications and data requirements as on modeling methods. An incremental model improvement approach toward advanced statewide models is recommended. Review findings also suggest model improvement activities should be justified by planning application needs. For statewide model improvement plans to be successful and financially sustainable, the return on model improvement investment needs to be demonstrated by timely applications that rely on improved model capabilities.     

Transit network design with allocation of green vehicles: A genetic algorithm approach

The use of fossil fuels in transportation generates harmful emissions that accounts for nearly half of the total pollutants in urban areas. Dealing with this issue, local authorities are dedicating specific efforts to seize the opportunity offered by new fuels and technological innovations in achieving a cleaner urban mobility. In fact, authorities are improving environmental performances of their public transport fleet by procuring cleaner vehicles, usually called low and zero emission vehicles (LEV and ZEV, respectively). Nevertheless there seems to be a lack of methodologies for supporting stakeholders in decisions related to the introduction of green vehicles, whose allocation should be performed since the network design process in order to optimize their available green capacity.In this paper, the problem of clean vehicle allocation in an existing public fleet is faced by introducing a method for solving the transit network design problem in a multimodal, demand elastic urban context dealing with the impacts deriving from transportation emissions.The solving procedure consists of a set of heuristics which includes a routine for route generation and a genetic algorithm for finding a sub-optimal set of routes with the associated frequencies.     

Scenario construction for transport planning

This paper argues that the types of uncertainty having the most serious consequences for transport planning are not captured by existing modelling procedures, which typically rely on a limited amount of sensitivity testing of plans based on extrapolative forecasts. It recommends as an alternative scenario‐based planning, an approach which is finding increasing application in private sector policy formation. The paper summarises the origins of scenario planning and the gradual broadening in its range of application, before going on to suggest methods of scenario construction suitable for transport planning applications. It considers the arguments for and against associating probabilities with scenarios and also addresses the problems of formulating transport strategies and evaluating them within the context of a scenario‐based approach to strategy choice.     

A traffic analysis zone definition: a new methodology and algorithm

This paper develops a comprehensive approach to the definition of transportation analysis zones (TAZ), and therein, presents a new methodology and algorithm for the definition of TAZ embedded in geographic information systems software, improves the base algorithm with several local algorithms, and comprehensively analyses the obtained results. The results obtained are then compared to these presently used in the transportation analysis process of the Lisbon Metropolitan Area. The proposed algorithm presents a new methodology for TAZ design based on a smoothed density surface of geocoded travel demand data. The algorithm aims to minimise the loss of information when moving from a continuous representation of the origin and destination of each trip to their discrete representations through zones, and focuses on the trade-off between the statistical precision, geographical error, and the percentage of intra-zonal trips of the resulting OD matrix. The results for the Lisbon Metropolitan Area case study suggest a significant improvement in OD matrix estimates compared to current transportation analysis practises based on administrative units.

The analysis of travel and emission impacts of travel demand management strategies using activity-based models

This paper demonstrates, tests and shows the value of activity-based travel demand models and household sample enumeration forecasting techniques in evaluating the transportation and air quality impacts of travel demand management strategies. Using data from the Portland, Oregon metropolitan area, three transportation policies were evaluated both individually and in combination: transit improvements, pricing, and telecommunications. The activity-based models used in this testing represents a significant improvement to today's "four-step" sequential model systems by providing a deeper insight into the individual decision making process in response to transportation policies. A wider range of impacts is predicted, and indirect effects as well as synergistic effects of such policies are taken into consideration. These models are capable of providing the information needed to improve the linkage of transportation models with emissions and air quality analysis methodologies by improving the prediction of variables that are important to accurately estimating emissions and air quality impacts of transportation actions.     

Transport models, philosophy and language

The aim of this paper is to encourage debate about the nature of transport modelling. It does so firstly by considering the underlying philosophies of science (apparently) adopted by transport modellers, over a period of more than 50 years, from the 1950s until the present day. The conclusion is that a new philosophy of science needs to be developed, which is more in tune with how transport modelling is actually carried out (as opposed to how early transport modellers thought it ought to be carried out). It is recommended that such a new philosophy perceives transport modelling as a linguistic activity within the overall context of transport planning, which is in turn considered as a communication process. The paper outlines three main approaches that could be taken in this respect, analysing transport models from metaphorical, narrative and aesthetic perspectives. Conclusions are drawn upon the possible future research directions that might follow from the analysis provided in the paper, emphasising the importance of bringing formal philosophical thinking into transport modelling research and practice.

Generating synthetic baseline populations from register data

The paper presents a population synthesiser based on the method of Iterative Proportional Fitting (IPF) algorithm developed for the new Danish national transport model system. The synthesiser is designed for large population matrices and allows target variables to be represented in several target constraints. As a result, constraints for the IPF are cross-linked, which makes it difficult to ensure consistency of targets in a forecast perspective. The paper proposes a new solution strategy to ensure internal consistency of the population targets in order to guarantee proper convergence of the IPF algorithm. The solution strategy consists in establishing a harmonisation process for the population targets, which combined with a linear programming approach, is applied to generate a consistent target representation. The model approach is implemented and tested on Danish administrative register data. A test on historical census data shows that a 2006 population could be predicted by a 1994 population with an overall percentage deviation of 5-6% given that targets were known. It is also indicated that the deviation is approximately a linear function of the length of the forecast period.     

Modelling demand in restricted parking zones

Multiple linear regression (MLR) and geographically weighted regression (GWR) models are used for estimating parking demand in areas with paid short stay parking systems. These models have been applied to the city of Santander (Cantabria, Spain) to check their goodness of fit and their predictive ability. The results show the main advantages and disadvantages of using GWR models. The technique proved to be useful in this case study because it offered a better fit and made better predictions in a scenario showing a certain degree of spatial heterogeneity unexplained by any of the variables introduced into the global model. However, the GWR model also presented situations of local correlation although this was considered moderate given the results provided by the variance inflation factors and the local condition indexes.