The current Thesis, attempts to study and analyse in depth personal underlying characteris-tics influencing parents in the mode choice procedure, regarding the transport mode their children will use for travelling to and from their school unit. The research aims to identify aspects of human behaviour regarding the execution and completion of school trips, while at the same time attempts to reveal the importance and significance parents attribute to spe-cific parameters that positively or negatively affect the mode choice decision.School transportation as well as other pupils’ extracurricular activities is an integral and im-portant part of society including many aspects such as knowledge attainment, socialization, identity building etc. Several definitions have be given to the school transportation, depend-ing on the way it is executed. In general, it is defined as the transfer of students to and from school, school events and activities.Investigating further deeper the wider concept of school transportation, it can be claimed that for a society, it is one of the most crucial, costly and responsible services provided to public, as it substantially ensures the children’s access to school. For that reason, Govern-ments and Local Authorities try to develop strategies that lead to a well-structured and or-ganized system for the management and control of the services provided.Travelling to school and back, forms mobility habits and travel behaviour aspects of students from a very young age, also adopted in their later life. The key players of the whole decision process are students’ parents, as in most cases they are the ones to decide how their chil-dren will complete everyday trips from their residence to the school unit and vice versa. Thus, parents' perceptions regarding the use of transport modes greatly influence the mobil-ity attitude students adopt from a very young age.The above-mentioned issues combined to the fact that the use of private vehicles for school trips has increased in recent decades against alternative transport modes (bicycle, walking, use of Public Transport System), pose an imperative need of investigating the parameters playing a catalytic role in the decision – making process. A further analysis of the motives behind the parents’ mode choice decision on school trips, may lead to solutions’ and strate-gies’ proposals which may reverse this trend in the near future, while it will also provide use-ful information to policy-makers, transport and spatial planners on how to overcome possi-ble barriers and difficulties in order to satisfactory cover all students’ future mobility needs.The basic stages the current research followed are:The first stage included the review of the corresponding bibliographic sources related to the research scope. It covered areas related to the existing school transportation institutional framework in Greece and international, issues related to statistic data regarding provided safety levels of a school transportation system and issues related to parents’ decision-making process regarding the mode choice for school travel. The combined critical review of the above mentioned issues, demarcated the problem, identified the research questions and de-termined the subject and purpose of this dissertation.The second stage included the design of the research, thus the design of an appropriate questionnaire in order to investigate the research questions according to the findings of the first stage. The research was conducted on parents of public Primary and High schools locat-ed in the wider area of the city of Thessaloniki in Greece. The sampling was carried out in full compliance with the sampling directives.The third stage included the process of conducting the survey, the collection of relevant data and the recording of this data in a relational database. The total period of data collection amounted approximately three months.In the fourth stage, the statistical analysis of the collected data took place including the de-scriptive statistical analysis, the Exploratory Factor Analysis (EFA) and the Confirmatory Fac-tor Analysis (CFA) which were the two main steps for the Structural Equation Model(SEM) development. SEM was structured focused on the main purpose of the current research, thus the identification of factors’ interrelations affecting the mode choice decision process re-garding the completion of school trips. Finally, three different Artificial Neural Networks (ANN) were constructed in order to identify the behavioural patterns of parents that lead to the prediction of specific decisions regarding students’ mode choice for students. The con-struction of ANNs was deemed necessary during the last stage of the present dissertation due to the ability of ANNs to generalize, thus the ability to be used in studies beyond those for which they were trained for.Regarding the literature review, the main conclusions are:The school transportation system in Greece presents common protocols with other European and international countries The alternative transportation as an option (cycling, walking school bus) even though it is successfully implemented in other countries it is not yet widespread in GreeceDespite the fact that the issue of school road safety is crucial, gaps are identified in the recording and provision of quantified traffic accident data regarding school trips both on local and national level.Safest school transport modes are found to be the school bus and the use of public transport systems, while a large percentage of injuries and deaths are linked with the use of alternative transport modes (walking, bicycling)Within the last thirty years, very significant reductions have been achieved in the percentages of traffic accidents in general and also in those involving children, main-ly due to policies and laws enforcedCritical parameters influencing parents in the mode choice decision process, are stu-dents’ age and gender, parents’ professional status, family income, car ownership in-dex, distance of residence from the school unit, route’s safety and parents' percep-tions regarding the use of different transport modes as well as mobility habits adopt-ed for daily trips completion.Regarding the most significant findings of the questionnaire survey on parents of Primary and High School students in the city of Thessaloniki, Greece, these are summarized below: Parents are very sceptical of allowing students to travel by bicycle to and from the school unitParents are captive of private vehicles use as they:a)consider that the prevailing traffic conditions on the city's road network are quite dangerous for students’ safety (traffic congestion, unsafe crossings, etc.),b)the existing Public Transport System provides inadequate services to passengers (unreliable itineraries, time delays, lack of comfortable conditions within the vehi-cles, etc.),c)there is a lack of appropriate and safe infrastructure for bicycle use.Differences are identified between students’ age categories regarding the transport mode used for a school trip. More specifically, High school students’ parents are in favour of walking as an alternative option, strengthening therefore the perception that older students travel more autonomouslyDifferences in students’ school trips are observed between the morning and the even-ing shift. Percentages of walking and use of school bus provided by the school unit, are increasing in the afternoon shift while a corresponding decrease is observed in the private vehicle use choice.The most important parameters for selecting a transport mode are safety, student’s age and the distance of the residence from the school unit, while the family income and travel cost are the least important.Regarding the Structural Equation Μodel developed for the needs of the present study, the most important conclusions emerged are:Parents’ perception regarding the use of bus seems to be affected by the safety pro-vided by the route from the stop to the school unit or the residence and vice versa.Parents’ perception regarding the use of private vehicle seems to influence their choice in favour of the private vehicle.Parents’ perception regarding the use of non-motorized transport modes (walking and bicycling) seems to have a positive effect on parents’ choice regarding the school trip completion. However, the positive attitude seems to lose its validity when par-ents take into account parameters related to the students’ age and safety as well as the travel distance and the time required to for the trip’s completion.A route from the student's residence to the school unit that is not safe, motivates parents to use private vehicle for the school trip.In addition, when a route is considered as safe, parameters such as time and travel distance, student’s age and parents’ working hours seem to lose their power as they are largely ignored by the parent.Regarding the development of Artificial Neural Networks, 3 different architectures were examined, including 38 inputs (answers to the questions of the research questionnaire) and one output, the mode choice. The architectures examined are:• Architecture A: 38 inputs, 50 nodes (neurons) in a hidden layer, and 1 output (38-50-1). The output consists of either three categories-classes (private vehicle, bus and non-motorized modes (walking/ bicycle), or two (motorized modes (private vehicle and bus, non-motorized modes (walking/ bicycle).• Architecture B: 38 inputs, 55 + 10 nodes (neurons) in two hidden layers and 1 output (38-55-10-1). The output consists of either three categories-classes (private vehicle, bus, non-motorized modes(walking / bicycle), or two (motorized modes (private vehicle and bus, non-motorized modes( walking / bicycle)).• Architecture C: 38 inputs, 65 + 15 nodes (neurons) in two hidden layers and 1 output (38-65-15-1). In this case, also the output consists of either three categories-(private ve-hicle, bus, non-motorized modes(walking / bicycle), or two (motorized modes (private vehicle and bus, non-motorized modes( walking / bicycle)).For each of the above architectures, the three options of the school trip included in the re-search questionnaire were examined:Current transport mode used for the morning school trip completion (residence-school unit) Current transport mode used for the morning school trip completion (school-residence)Preferred transport modeEach ANN was trained 10 times. For each of three parent choices (preferred transport mode, current transport mode used for the morning school trip completion and current transport mode used for the morning school trip completion) 10 experimental tests (simulations) were preformed to confirm the stability of the classifier. The total number of finally trained ANNs was 180.The three categories ANN analysis (ANN 38-50-1, ANN 38-55-10-1, ANN 38-65-15-1) under three options of school travel-(private vehicle, bus, non-motorized modes(walking / bicy-cle),led to the following conclusions:A hidden layers increase from the one hidden level architecture (ANN 38-50-1), to the two hidden layers (ANN 38-55-10-1 and ANN 38-65-15-1) architectures does not seem to have any particular significant effect on the optimization of the ANNs predictive capacity as the percentages of successful forecasts show minor differ-ences between them.ANNs systematically overestimate the non-motorized mode choice (walk-ing/bicycle), compared to private vehicle or bus choices.Trained ANNs systematically underestimate the use of bus, as forecast rates are clearly lower than the percentages provided by the questionnaire survey in both the preferable and the current transport mode used for the morning and the af-ternoon school trip completion.Regarding the transport mode used, a difference is found in the predictive capac-ity of the ANNs refer to the use private vehicle between the morning and the af-ternoon school trip. While in the morning trip (residence– school unit) the ANN predicted higher rates of private vehicle use than the actual rates (as they emerged from the questionnaire survey), in the afternoon trip (school unit-residence) the exact opposite result has been noticed, therefore the ANNs under-estimate the selection and use of private vehicle.Finally, the three categories ANN analysis (ANN 38-50-1, ANN 38-55-10-1, ANN 38-65-15-1) under two options of school travel-(motorized modes (private vehicle and bus), non-motorized modes(walking / bicycle), led to the following conclusions:A hidden layers increase from the one hidden level architecture (ANN 38-50-1), to the two hidden layers (ANN 38-55-10-1 and ANN 38-65-15-1) architectures does not seem to have any particular significant effect on the optimization of the ANNs predictive capacity as the percentages of successful forecasts show minor differ-ences between them.The vast majority of ANNs overestimate the non-motorized travel choices (walk-ing/bicycle) over the motorized choices (private vehicle and bus).The same finding applies in the case of the afternoon transport mode choice (school unit - residence), while in the morning case differences have been no-ticed.In conclusion, the results of this Thesis provide a methodology of predicting parents’ mode choice on school trips through the recognition of behaviour patterns leading to specific deci-sions. The Artificial Neural Networks training based on input data extracted from reliable and documented questionnaire surveys, produce satisfactory predictions regarding the use of specific transport mode for the school trip completion providing therefore an important and easy-to-use tool in the scientific community toolbox as well as in the planning and im-plementation of sustainable urban mobility policies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.