Determination of the influence factors on household vehicle ownership patterns in Phnom Penh using statistical and machine learning methods

Ha, Tran Vinh; Asada, Takumi; Arimura, Mikiharu

doi:10.1016/j.jtrangeo.2019.05.015

Cited by 15 publications

(13 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The findings in this study throws more light on vehicle ownership in the context of a small to medium urban city in Ghana which is mainly influence by the respondents travel characteristics and average monthly income. This finding is largely in agreements with the studies by Ha et al [ 33 ], Kaewwichiann et al [ 34 ] and Paredes et al [ 32 ], however, with regards to the best performing ML algorithm, the finding in our study produced different results which might be as a result of the difference in the data structure and the methods used to find the best learning features for each algorithm, highlighting the case-specific nature of the application of the ML algorithms to vehicle ownership studies.…”

Section: 0 Conclusionsupporting

confidence: 94%

“…Their results show that ML algorithms outperform the MNL model with RF having the highest performance in terms of accuracy, noting that ML and MNL are not interchangeable but complementary in terms of vehicle ownership modeling. Ha et al [ 33 ] also examined the feature impact level of ANN, RF and MNL to predict vehicle ownership levels in the city of Phnom Penh. Their results indicated that household income was the most prominent feature affecting vehicle ownership in Phnom Penh and that the RF produced the highest accuracy in terms of prediction.…”

Section: 0 Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling vehicle ownership with machine learning techniques in the Greater Tamale Area, Ghana

2021

View full text Add to dashboard Cite

Vehicle ownership modeling and prediction is a crucial task in the transportation planning processes which, traditionally, uses statistical models in the modeling process. However, with the advancement in computing power of computers and Artificial Intelligence, Machine Learning (ML) algorithms are becoming an alternative or a complement to the statistical models in modeling the transportation planning processes. Although the application of ML algorithms to the transportation planning processes—like mode choice, and traffic forecasting and demand modeling—have received much attention in research and abound in literature, scanty attention is paid to its application to vehicle ownership modeling especially in the context of small to medium cities in developing countries. Therefore, this study attempts to fill this gap by modeling vehicle ownership in the Greater Tamale Area (GTA), a typically small to medium city in Ghana. Using a cross sectional survey of formal sectors workers, data was collected between June–August 2018. The study applied nine different ML classification algorithms to the dataset using 10-fold cross-validation technique/s and the Cohen-Kappa static/statistic to evaluate the predictive performance of each of the algorithms, and the Permutation Feature Importance to examine the features that contribute significantly to the prediction of vehicle ownership in GTA. The results showed that Linear Support Vector Classification (LinearSVC) classifier performed well in comparison with the other classifiers with regards to the overall predictive ability of the classifiers. In terms of class predictions, K- Nearest Neighbors (KNN) classifier performs well for no-vehicle class whiles Linear Support Vector Classification (LinearSVC) and GaussianNB classifiers performs well for motorcycle ownership. LinearSVC and Logistic Regression classifiers performed well on the car ownership class. Also, the results indicated that travel mode choice, average monthly income, average travel distance to workplace, average monthly expenditure on transport, duration of travel to workplace, occupational rank, age, household size and marital status were significant in predicting vehicle ownership for most of the classifiers. These findings could help policies makers carve out strategies that would reduce vehicle ownership but improve personal mobility.

show abstract

Section: 0 Conclusionsupporting

confidence: 94%

Section: 0 Introductionmentioning

confidence: 99%

Modeling vehicle ownership with machine learning techniques in the Greater Tamale Area, Ghana

2021

View full text Add to dashboard Cite

show abstract

“…For example, Loo et al [49] revealed that SNs play a vital role in people's mobility decisions as to whether to use a private vehicle or public transport mode. In line with Loo's study, Ha et al [50] indicate that Cambodian people decide to own private vehicles because of disconnected streets, long distances to reach destinations, and unreliable public transport services.…”

Section: Comparison Of Dsns and Wsns Across The Ascnmentioning

confidence: 57%

Analysis of Urban Drivable and Walkable Street Networks of the ASEAN Smart Cities Network

Zhao

Yen

Bailey

et al. 2019

IJGI

View full text Add to dashboard Cite

Making transport systems sustainable is a topic that has attracted the attention of many researchers and urban planners. The Association of Southeast Asian Nations (ASEAN) Smart Cities Network (ASCN) was initiated to develop a sustainable transport system in the ASEAN countries. A comprehensive understanding of street networks (SNs) can contribute significantly to the achievement of this initiative. Therefore, this paper measured and compared characteristics of drivable street networks (DSNs) and walkable street networks (WSNs) of the 26 ASCN pilot cities by applying multiple network metrics. The OSMnx tool was used to download and analyse WSNs and DSNs from the OpenStreetMap. The findings present the topological and geometric characteristics of WSNs and DSNs that are diverse and characterized by different factors. The cities with orthogonal street grids, high street density, intersection density, and fewer cul-de-sacs have good accessibility to reach destinations. In contrast, some other cities have more curvilinear and circuitous SNs with many missing links to other streets, which in turn are prone to traffic disruption. The study highlights the important features of SNs that have significant implications for future designs of SNs in the ASCN whose goal is to make cities smart and liveable for ASEAN members.

show abstract

“…The decision to own a private vehicle and the type of vehicle depends on different HH characteristics, such as its income, size, the number of license holders, composition in full-time workers and children, education level, gender and age [1], [2]. Ha, et al (2019), using important variable ranking methods as Multi-nominal Logit model, Neural Networks and Random Forests, found that income is the most potent variable influence on motorisation among other HH characteristics [5]. Schievelbein et al (2016) surveyed India to predict the type of vehicle, including motorised two wheels and four wheels that a HH will own by using a Multi-Nominal Logit model (MNL).…”

Section: Literature Reviewmentioning

confidence: 99%

“…Moreover, it was found that high-income HHs tend to own luxury vehicles rather than own more vehicles [2]. In Phnom Penh in 2019, Ha et al has applied the MNL, neural networks, and random forest and found that the presence of children in a HH appears to be another factor that determines the type of vehicles and results in a higher level of mobility, convenience and safety [5]. Kim et al has used MNL and found that HHs in the United States commonly choose vans when they have more children under 8, or have older primary drivers [8].…”

Section: Literature Reviewmentioning

confidence: 99%

Analysis of Vehicle Ownership Attributes in Western Province, Sri Lanka

Dilini

Amalan

Kumarage

2021

J. S. Asian Logistics Transport

View full text Add to dashboard Cite

This paper develops a mathematical model to predict motor vehicle ownership based on household (HH) characteristics. The model is tested using household visit surveys in the Western Province of Sri Lanka (CoMTrans, 2014). The province which has the country's highest population density (1,600/km 2 ) and road density (0.9 km/km 2 ) as well as a motor vehicle ownership of 206 vehicles per 1,000 people. The modelling is disaggregated into motorcycles, three-wheelers, vans, and cars (including jeeps and pick-ups). The motor vehicle fleet comprises 51% motorcycles, 20.2% threewheelers, 6.7% vans, and 17.7% cars apart from commercial vehicles. The purchasing cost of motor vehicles in Sri Lanka varies widely due to different taxes imposed at importation. A binary logistic regression with cross-validation statistical theories was used to predict the HH ownership of different vehicles, based on an income-based testing scenario for determining a HH's likelihood of owning a particular type of vehicle.Motorcycles, three-wheelers, vans, and cars listed in ascending order of cost of ownership and operation were tested against the characteristics of 35,850 HHs using R, a software analytical tool. The analysis found that private vehicle ownership depends on attributes of a HH, such as its size, average monthly income, and the percentage of workers, school and kindergarten children, and males in that HH.

show abstract

Determination of the influence factors on household vehicle ownership patterns in Phnom Penh using statistical and machine learning methods

Cited by 15 publications

References 55 publications

Modeling vehicle ownership with machine learning techniques in the Greater Tamale Area, Ghana

Modeling vehicle ownership with machine learning techniques in the Greater Tamale Area, Ghana

Analysis of Urban Drivable and Walkable Street Networks of the ASEAN Smart Cities Network

Analysis of Vehicle Ownership Attributes in Western Province, Sri Lanka

Contact Info

Product

Resources

About