A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information
Discrete choice modeling of travel modes is an essential part of traffic planning and management. Thus far, this field has been dominated by multinomial logit (MNL) models with a linear utility specification. However, deep neural networks (DNNs), owing to their powerful capacity of nonlinear fitting...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Journal of Advanced Transportation |
| Online Access: | http://dx.doi.org/10.1155/2020/5364252 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832547378586976256 |
|---|---|
| author | Zhengchao Zhang Congyuan Ji Yineng Wang Yanni Yang |
| author_facet | Zhengchao Zhang Congyuan Ji Yineng Wang Yanni Yang |
| author_sort | Zhengchao Zhang |
| collection | DOAJ |
| description | Discrete choice modeling of travel modes is an essential part of traffic planning and management. Thus far, this field has been dominated by multinomial logit (MNL) models with a linear utility specification. However, deep neural networks (DNNs), owing to their powerful capacity of nonlinear fitting, are now rapidly replacing these models. This is because, by using DNNs, mode choice can be assimilated with the classification problems within the machine learning community. This article proposes a newly designed DNN framework for traffic mode choice in the style of two hidden layers. First, a local-connected layer automatically extracts an effective utility specification from the available data, and then, a fully connected layer augments the feature representation. Validated by a practical city-wide multimodal traffic dataset in Beijing, our model significantly outperforms the random utility models and simple fully connected neural network in terms of the prediction accuracy. Besides the comparison of the predictive power, we also present the interpretability of the proposed model. |
| format | Article |
| id | doaj-art-a2045a5bb0df4b69a353ae93820ba19b |
| institution | Kabale University |
| issn | 0197-6729 2042-3195 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Journal of Advanced Transportation |
| spelling | doaj-art-a2045a5bb0df4b69a353ae93820ba19b2025-02-03T06:44:57ZengWileyJournal of Advanced Transportation0197-67292042-31952020-01-01202010.1155/2020/53642525364252A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility InformationZhengchao Zhang0Congyuan Ji1Yineng Wang2Yanni Yang3Department of Civil Engineering, Tsinghua University, Beijing 100084, ChinaDepartment of Industrial Engineering, Tsinghua University, Beijing 100084, ChinaDepartment of Industrial Engineering, Tsinghua University, Beijing 100084, ChinaSchool of Management and Engineering, Capital University of Economics and Business, Beijing 100070, ChinaDiscrete choice modeling of travel modes is an essential part of traffic planning and management. Thus far, this field has been dominated by multinomial logit (MNL) models with a linear utility specification. However, deep neural networks (DNNs), owing to their powerful capacity of nonlinear fitting, are now rapidly replacing these models. This is because, by using DNNs, mode choice can be assimilated with the classification problems within the machine learning community. This article proposes a newly designed DNN framework for traffic mode choice in the style of two hidden layers. First, a local-connected layer automatically extracts an effective utility specification from the available data, and then, a fully connected layer augments the feature representation. Validated by a practical city-wide multimodal traffic dataset in Beijing, our model significantly outperforms the random utility models and simple fully connected neural network in terms of the prediction accuracy. Besides the comparison of the predictive power, we also present the interpretability of the proposed model.http://dx.doi.org/10.1155/2020/5364252 |
| spellingShingle | Zhengchao Zhang Congyuan Ji Yineng Wang Yanni Yang A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information Journal of Advanced Transportation |
| title | A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information |
| title_full | A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information |
| title_fullStr | A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information |
| title_full_unstemmed | A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information |
| title_short | A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information |
| title_sort | customized deep neural network approach to investigate travel mode choice with interpretable utility information |
| url | http://dx.doi.org/10.1155/2020/5364252 |
| work_keys_str_mv | AT zhengchaozhang acustomizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT congyuanji acustomizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT yinengwang acustomizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT yanniyang acustomizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT zhengchaozhang customizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT congyuanji customizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT yinengwang customizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation AT yanniyang customizeddeepneuralnetworkapproachtoinvestigatetravelmodechoicewithinterpretableutilityinformation |