Search

• Used statistical learning theory to evaluate DNNs in choice analysis. • Operationalized DNN interpretability by using the choice probability functions. • Provided a tight upper bound on the estimation error of DNNs. • Conducted experiments to identify when DNNs outperform classical models. • DNNs can be more predictive and interpretable than BNL and MNL models. Although researchers increasingly use deep neural networks (DNN) to analyze individual choices, overfitting and interpretability issues remain obstacles in theory and practice. This study presents a statistical learning theoretical framework to examine the tradeoff between estimation and approximation errors, and between the quality of prediction and of interpretation. It provides an upper bound on the estimation error of the prediction quality in DNN, measured by zero-one and log losses, shedding light on why DNN models do not overfit. It proposes a metric for interpretation quality by formulating a function approximation loss that measures the difference between true and estimated choice probability functions. It argues that the binary logit (BNL) and multinomial logit (MNL) models are the specific cases of DNNs, since the latter always has smaller approximation errors. We explore the relative performance of DNN and classical choice models through three simulation scenarios comparing DNN, BNL, and binary mixed logit models (BXL), as well as one experiment comparing DNN to BNL, BXL, MNL, and mixed logit (MXL) in analyzing the choice of trip purposes based on the National Household Travel Survey 2017. The results indicate that DNN can be used for choice analysis beyond the current practice of demand forecasting because it has the inherent utility interpretation and the power of automatically learning utility specification. Our results suggest DNN outperforms BNL, BXL, MNL, and MXL models in both prediction and interpretation when the sample size is large (≥ O (10 4)), the input dimension is high, or the true data generating process is complex, while performing worse when the opposite is true. DNN outperforms BNL and BXL in zero-one, log, and approximation losses for most of the experiments, and the larger sample size leads to greater incremental value of using DNN over classical discrete choice models. Overall, this study introduces the statistical learning theory as a new foundation for high-dimensional data, complex statistical models, and non-asymptotic data regimes in choice analysis, and the experiments show the effective prediction and interpretation of DNN for its applications to policy and behavioral analysis. [ABSTRACT FROM AUTHOR]

• Created a TB-ResNet framework to synergize theory- and data-driven methods. • Exemplified the synergy by combing discrete choice models and deep neural networks. • Adapted TB-ResNets for multinomial logit, risk, and time preferences. • Empirically tested the TB-ResNets for travel behavior analysis. • Illustrated TB-ResNets' strength in prediction, interpretation, and robustness. Researchers often treat data-driven and theory-driven models as two disparate or even conflicting methods in travel behavior analysis. However, the two methods are highly complementary because data-driven methods are more predictive but less interpretable and robust, while theory-driven methods are more interpretable and robust but less predictive. Using their complementary nature, this study designs a theory-based residual neural network (TB-ResNet) framework, which synergizes discrete choice models (DCMs) and deep neural networks (DNNs) based on their shared utility interpretation. The TB-ResNet framework is simple, as it uses a (δ , 1- δ) weighting to take advantage of DCMs' simplicity and DNNs' richness, and to prevent underfitting from the DCMs and overfitting from the DNNs. This framework is also flexible: three instances of TB-ResNets are designed based on multinomial logit model (MNL-ResNets), prospect theory (PT-ResNets), and hyperbolic discounting (HD-ResNets), which are tested on three data sets. Compared to pure DCMs, the TB-ResNets provide greater prediction accuracy and reveal a richer set of behavioral mechanisms owing to the utility function augmented by the DNN component in the TB-ResNets. Compared to pure DNNs, the TB-ResNets can modestly improve prediction and significantly improve interpretation and robustness, because the DCM component in the TB-ResNets stabilizes the utility functions and input gradients. Overall, this study demonstrates that it is both feasible and desirable to synergize DCMs and DNNs by combining their utility specifications under a TB-ResNet framework. Although some limitations remain, this TB-ResNet framework is an important first step to create mutual benefits between DCMs and DNNs for travel behavior modeling, with joint improvement in prediction, interpretation, and robustness. [ABSTRACT FROM AUTHOR]

Classical demand modeling analyzes travel behavior using only low-dimensional numeric data (i.e. sociodemographics and travel attributes) but not high-dimensional urban imagery. However, travel behavior depends on the factors represented by both numeric data and urban imagery, thus necessitating a synergetic framework to combine them. This study creates a theoretical framework of deep hybrid models consisting of a mixing operator and a behavioral predictor, thus integrating the numeric and imagery data for travel behavior analysis. Empirically, this framework is applied to analyze travel mode choice using the Chicago MyDailyTravel Survey as the numeric inputs and the satellite images as the imagery inputs. We found that deep hybrid models significantly outperform both classical demand models and deep learning models in predicting aggregate and disaggregate travel behavior. The deep hybrid models can reveal spatial clusters with meaningful sociodemographic associations in the latent space. The models can also generate new satellite images that do not exist in reality and compute the corresponding economic information, such as substitution patterns and social welfare. Overall, the deep hybrid models demonstrate the complementarity between the low-dimensional numeric and high-dimensional imagery data and between the traditional demand modeling and recent deep learning. They enrich the family of hybrid demand models by using deep architecture as the latent space and enabling researchers to conduct associative analysis for sociodemographics, travel decisions, and generated satellite imagery. Future research could address the limitations in interpretability, robustness, and transferability, and propose new methods to further enrich the deep hybrid models. • Designing deep hybrid models to integrate numeric and imagery data. • Deep hybrid models outperform demand models and deep learning in demand prediction. • Deep hybrid models have high-dimensional latent space for image processing and generation. • The latent space is spatially and socially meaningful. • Deep hybrid models can generate new urban imagery and derive economic information. [ABSTRACT FROM AUTHOR]

• Use behavioral knowledge to design a new DNN architecture with alternative-specific utility (ASU-DNN). • ASU-DNN provides a more regular substitution pattern of travel mode choices. • ASU-DNN improves both the predictive power and interpretability. • Behavioral knowledge can function as an effective domain-knowledge-based regularization. Whereas deep neural network (DNN) is increasingly applied to choice analysis, it is challenging to reconcile domain-specific behavioral knowledge with generic-purpose DNN, to improve DNN's interpretability and predictive power, and to identify effective regularization methods for specific tasks. To address these challenges, this study demonstrates the use of behavioral knowledge for designing a particular DNN architecture with alternative-specific utility functions (ASU-DNN) and thereby improving both the predictive power and interpretability. Unlike a fully connected DNN (F-DNN), which computes the utility value of an alternative k by using the attributes of all the alternatives, ASU-DNN computes it by using only k 's own attributes. Theoretically, ASU-DNN can substantially reduce the estimation error of F-DNN because of its lighter architecture and sparser connectivity, although the constraint of alternative-specific utility can cause ASU-DNN to exhibit a larger approximation error. Empirically, ASU-DNN has 2–3% higher prediction accuracy than F-DNN over the whole hyperparameter space in a private dataset collected in Singapore and a public dataset available in the R mlogit package. The alternative-specific connectivity is associated with the independence of irrelevant alternative (IIA) constraint, which as a domain-knowledge-based regularization method is more effective than the most popular generic-purpose explicit and implicit regularization methods and architectural hyperparameters. ASU-DNN provides a more regular substitution pattern of travel mode choices than F-DNN does, rendering ASU-DNN more interpretable. The comparison between ASU-DNN and F-DNN also aids in testing behavioral knowledge. Our results reveal that individuals are more likely to compute utility by using an alternative's own attributes, supporting the long-standing practice in choice modeling. Overall, this study demonstrates that behavioral knowledge can guide the architecture design of DNN, function as an effective domain-knowledge-based regularization method, and improve both the interpretability and predictive power of DNN in choice analysis. Future studies can explore the generalizability of ASU-DNN and other possibilities of using utility theory to design DNN architectures. [ABSTRACT FROM AUTHOR]

Despite an increasingly large body of research that focuses on the potential demand for autonomous vehicles (AVs), risk preference is an understudied factor. Given that AV technology and how it will interact with the evolving mobility system are highly risky, this lack of research on risk preference is a critical gap in current understanding. By using a stated preference survey of 1142 individuals from Singapore, this study achieves three objectives. First, it develops one measure of psychometric risk preference and operationalizes prospect theory to create two economic risk preference parameters. Second, it examines how these psychometric and economic risk preferences are associated with socioeconomic variables. Third, it analyzes how risk preference influences the mode choice of AVs. The study finds that risk preference parameters are significantly associated with socioeconomic variables: the elderly, poor, females, and unemployed Singaporeans appear more risk-averse and tend to overestimate small probabilities of losses. Furthermore, all three risk preference parameters contribute to the prediction of AV adoption. These modeling results have policy implications at both the aggregate and disaggregate levels. At the aggregate level, people misperceive probabilities, are overall risk-averse, and hence under-consume AVs relative to the social optimum. At the disaggregate level, the elderly, poor, female, and unemployed are more risk-averse and thus are less likely to adopt AVs. These results suggest that it might be valuable for governments to implement policies to encourage technology adoption, particularly for disadvantaged social groups, although caution remains due to uncertainty in the long-term effects of AVs. Individualized risk preference parameters could also inform how to design regulations, safety standards, and liability allocations of AVs since many regulations are essentially mechanisms for risk allocation. One limitation of the paper is that risk preference is measured and modeled only as individual-specific but not alternative-specific variables. Future studies should examine the relationship between the multiple components of risk preference and the multiple risky aspects of AVs. [ABSTRACT FROM AUTHOR]

Lotteries and auctions are common ways of allocating public resources, but they have rarely been used simultaneously in urban transportation policies. This paper presents a unique policy experiment in Guangzhou, China, where lotteries and auctions are used in conjunction to allocate vehicle licenses. Guangzhou introduced vehicle license regulations to control the monthly quota of local automobile growth in 2012. To obtain a license, residents are required to choose between the lottery and auction method. Since the introduction of the regulations, there has been heated debates on the distributional effects of lotteries and auctions; however, the debates have not been grounded in empirical studies. We analyze the distributional effects of such mixed mode of resource allocation in a positive manner based on individual behavioral choices. We conducted a survey in January 2016 (n = 1000 people ∗ 12 months), and used mixed logit models to analyze how socio-economic status, including income and household automobile ownership, determined people’s choices among lottery, auction, and non-participation alternatives. We find that income increased participation, but did not influence non-car owners’ choices between lotteries and auctions, which contrasts with the common notion that lotteries benefit the poor. Additionally, the positive impact of car ownership on participation indicates a car-dependent trajectory for automobile growth. The significant socio-economic differentiators between lotteries and auctions were age, gender, and education. Proxies of mobility needs were insignificant overall. The program attributes had a much larger impact than all other variables—people were more likely to choose lotteries with higher winning rates and more participants and more likely to choose auctions with higher prices and more participants. We concluded that for those who participated, the choice between lotteries and auctions did not depend on their income or mobility needs but, rather, the probability of winning plates and the opportunity for speculation. [ABSTRACT FROM AUTHOR]

In this paper, a theory is developed for analysis of steel-concrete composite beams with interfacial slip and shear deformation in steel considered. Closed-form solutions are derived for simply supported composite beam under uniform and mid-span concentrated load respectively. Examples are given to compute various quantities of beams. Comparing the results calculated by closed-form solutions with those by ANSYS, good agreements are found. Finally, an explicit formula for the equivalent bending stiffness of the composite beams is found by which the deflections of the composite beams can be calculated as if they were common Bernoulli beams. [ABSTRACT FROM AUTHOR]

Methyl parathion (MP) as an organophosphorus pesticide has been used in the control of agricultural pests and diseases. Due to its high toxicity and persistence in the environment, MP may pose threat to human health when it is released into environmental water. For MP treatment, people have found that oxidative degradation of MP may generate some intermediates which are more toxic than MP itself, such as methyl paraoxon. Herein, we proposed a new method of applying dielectric barrier discharge (DBD) non-thermal plasma technology to treat MP in aqueous solution, and investigated the influences of different gases, pH value, discharge voltage/power, and main active species on the MP removal efficiency. In particular, the safety of DBD treatment was concerned with analysis of the biological toxicity of the byproducts from the DBD oxidation, and the DBD-induced degradation together with the involved mechanism was explored therein. The results showed that the production of toxic intermediates could be effectively suppressed or avoided under certain treatment conditions. As such, this work demonstrates that the proper application of DBD plasma technology with necessary caution can detoxify methyl parathion effectively, and also provides a practical guide for low-temperature plasma application in treatment of various organophosphorus pesticides in agricultural wastewater. [Display omitted] • Efficient detoxification of methyl parathion by dielectric barrier discharge (DBD) plasma was explored. • The comprehensive DBD induced degradation pathways and mechanisms were scrutinized. • Biotoxicity evaluation on various DBD degradation intermediates was provided. • This work may give a guide to efficient and safe application of plasma in pesticide treatment. [ABSTRACT FROM AUTHOR]

Since its inception, the choice modelling field has been dominated by theory-driven modelling approaches. Machine learning offers an alternative data-driven approach for modelling choice behaviour and is increasingly drawing interest in our field. Cross-pollination of machine learning models, techniques and practices could help overcome problems and limitations encountered in the current theory-driven modelling paradigm, such as subjective labour-intensive search processes for model selection, and the inability to work with text and image data. However, despite the potential benefits of using the advances of machine learning to improve choice modelling practices, the choice modelling field has been hesitant to embrace machine learning. This discussion paper aims to consolidate knowledge on the use of machine learning models, techniques and practices for choice modelling, and discuss their potential. Thereby, we hope not only to make the case that further integration of machine learning in choice modelling is beneficial, but also to further facilitate it. To this end, we clarify the similarities and differences between the two modelling paradigms; we review the use of machine learning for choice modelling; and we explore areas of opportunities for embracing machine learning models and techniques to improve our practices. To conclude this discussion paper, we put forward a set of research questions which must be addressed to better understand if and how machine learning can benefit choice modelling. • Clarifies the similarities and differences between theory and data-driven paradigms. • Reviews the use of machine learning for choice modelling. • Explores opportunities for embracing machine learning to benefit choice modelling. • Puts forward research agenda. [ABSTRACT FROM AUTHOR]

Antibiotics are abused and discharged into environmental water, posing a constant potential threat to ecosystem. It is still a challenge to treat the wastewater containing various antibiotics. While many studies have reported the plasma treatment of different antibiotics, how to treat multiple antibiotics simultaneously with high efficiency still remains an unsolved problem. In this work, we employed atmospheric-pressure dielectric barrier discharge (DBD) to treat levofloxacin and sulfadiazine as two typical antibiotics in water, and investigated the treatment efficiency and the involved mechanism. The experimental results indicated that under proper conditions, the total antibiotics removal efficiency could be enhanced compared to separate single antibiotic treatment. The contributions from the plasma-induced reactive oxygen/nitrogen species (RONS) were examined, showing that ·OH played a major role in levofloxacin degradation, while ozone and peroxynitrite also played a certain role in sulfadiazine degradation. The bio-toxicity evaluation for the plasma treatment of levofloxacin and sulfadiazine was also provided, showing that the DBD degradation products were harmless to the ecological environment. As such, this work may not only provide a practical solution to treatment of wastewater containing multiple antibiotics, but also give the insights into the mechanism for the efficient DBD treatment of mixed antibiotics in wastewater. [Display omitted] • Dielectric barrier discharge (DBD) plasma was effective for removing levofloxacin and sulfadiazine in water. • The efficiency for the mixed antibiotics treatment was enhanced compared to that of single antibiotics treatment. • The reactive oxygen/nitrogen species (RONS) played an important role in the removal of mixed antibiotics. • The formation of nitrate in Air-DBD facilitated the removal of mixed antibiotics. • The biotoxicity evaluation confirmed the safety of NTP treatment of mixed antibiotics. [ABSTRACT FROM AUTHOR]

• Investigated computational fairness in travel behavior modeling. • Operationalized computational fairness by equality of opportunity. • Revealed prediction disparity for race, income, gender, health and region. • Adopted a bias mitigation method to improve fairness in travel behavior prediction. • Demonstrated the accuracy-fairness tradeoff. Although researchers increasingly adopt machine learning to model travel behavior, they predominantly focus on prediction accuracy, ignoring the ethical challenges embedded in machine learning algorithms. This study introduces an important missing dimension - computational fairness - to travel behavior analysis. It highlights the accuracy-fairness tradeoff instead of the single dimensional focus on prediction accuracy in the contexts of deep neural network (DNN) and discrete choice models (DCM). We first operationalize computational fairness by equality of opportunity , then differentiate between the bias inherent in data and the bias introduced by modeling. The models inheriting the inherent biases can risk perpetuating the existing inequality in the data structure, and the biases in modeling can further exacerbate it. We then demonstrate the prediction disparities in travel behavior modeling using the 2017 National Household Travel Survey (NHTS) and the 2018–2019 My Daily Travel Survey in Chicago. Empirically, DNN and DCM reveal consistent prediction disparities across multiple social groups: both over-predict the false negative rate of frequent driving for the ethnic minorities, the low-income and the disabled populations, and falsely predict a higher travel burden of the socially disadvantaged groups and the rural populations than reality. Comparing DNN with DCM, we find that DNN can outperform DCM in prediction disparities because of DNN's smaller misspecification error. To mitigate prediction disparities, this study introduces an absolute correlation regularization method, which is evaluated with synthetic and real-world data. The results demonstrate the prevalence of prediction disparities in travel behavior modeling, and the disparities still persist regarding a variety of model specifics such as the number of DNN layers, batch size and weight initialization. Since these prediction disparities can exacerbate social inequity if prediction results without fairness adjustment are used for transportation policy making, we advocate for careful consideration of the fairness problem in travel behavior modeling, and the use of bias mitigation algorithms for fair transport decisions. [ABSTRACT FROM AUTHOR]

• The interaction between the injured brain and distant GUT was explored. • Microbiota and metabolites changes after TBI were investigated. • The association between the differential microbiota and metabolites were analyzed. • Insights were provided to diagnose and treat TBI, using the brain gut axis theory. Traumatic Brain Injury (TBI) present a significant burden to global health. Close association and mutual regulation exist between the brain and gut microbiota. In addition, metabolites may play an important role as intermediary mediators of the brain and gut microbiota. Consequently, the study sought to investigate the alterations in gut microbiota and metabolites after TBI and conducted a comprehensive analysis of the correlation between gut microbiota and metabolites after TBI in mice. Changes in intestinal microbiota and metabolites in mice after moderate or severe traumatic brain injury were detected through 16S rDNA sequencing and the non-target LC-MS technology. Additionally, Pearson correlation analysis was used to explore the association between the microbiota and metabolites. TBI was able to change the composition of intestinal microbiota, resulting to a decrease in microbial diversity in the intestinal tract (sham vs sTBI: 8.35 ± 0.12 vs 7.71 ± 0.5, p < 0.01; sTBI vs mTBI: 7.71 ± 0.5 vs 8.25 ± 0.34, p < 0.05). The results also showed that TBI could change the types and abundance of metabolites (723 in mTBI and sham groups; 1221 in sTBI and sham groups; 324 in mTBI and sTBI groups). Moreover, some of the altered gut metabolites were significantly correlated with part of the altered gut microbes after TBI. TBI significantly changed intestinal microbiota as well as metabolites. Some of the altered microbiota and metabolites had a significant association. The results from this study provide information that paves way for future studies utilizing the brain gut axis theory in the diagnosis and treatment of TBI. [ABSTRACT FROM AUTHOR]

It is an enduring question how to combine revealed preference (RP) and stated preference (SP) data to analyze individual choices. While the nested logit (NL) model is the classical way to address the question, this study presents multitask learning deep neural networks (MTLDNNs) as an alternative framework, and discusses its theoretical foundation, empirical performance, and behavioral intuition. We first demonstrate that the MTLDNNs are theoretically more general than the NL models because of MTLDNNs' automatic feature learning, flexible regularizations, and diverse architectures. By analyzing the adoption of autonomous vehicles (AVs), we illustrate that the MTLDNNs outperform the NL models in terms of prediction accuracy but underperform in terms of cross-entropy losses. To interpret the MTLDNNs, we compute the elasticities and visualize the relationship between choice probabilities and input variables. The MTLDNNs reveal that AVs mainly substitute driving and ride hailing, and that the variables specific to AVs are more important than the socio-economic variables in determining AV adoption. Overall, this work demonstrates that MTLDNNs are theoretically appealing in leveraging the information shared by RP and SP and capable of revealing meaningful behavioral patterns, although its performance gain over the classical NL model is still limited. To improve upon this work, future studies can investigate the inconsistency between prediction accuracy and cross-entropy losses, novel MTLDNN architectures, regularization design for the RP-SP question, MTLDNN applications to other choice scenarios, and deeper theoretical connections between choice models and the MTLDNN framework. [ABSTRACT FROM AUTHOR]

• Extract economic information from DNN for choice analysis. • Introduce both function-based and gradient-based interpretations. • Highlight three challenges associated with the automatic learning capacity of DNN. • Compare economic information from DNN with those from discrete choice models. • Economic information aggregated either over trainings or population is more reliable. While deep neural networks (DNNs) have been increasingly applied to choice analysis showing high predictive power, it is unclear to what extent researchers can interpret economic information from DNNs. This paper demonstrates that DNNs can provide economic information as complete as classical discrete choice models (DCMs). The economic information from DNNs includes choice predictions, choice probabilities, market shares, substitution patterns of alternatives, social welfare, probability derivatives, elasticities, marginal rates of substitution, and heterogeneous values of time. Unlike DCMs, DNNs can automatically learn utility functions and reveal behavioral patterns that are not prespecified by domain experts, particularly when the sample size is large. However, the economic information obtained from DNNs can be unreliable when the sample size is small, because of three challenges associated with the automatic learning capacity: high sensitivity to hyperparameters, model non-identification, and local irregularity. The first challenge is related to the statistical challenge of balancing approximation and estimation errors of DNNs, the second to the optimization challenge of identifying the global optimum in the DNN training, and the third to the robustness challenge of mitigating locally irregular patterns of estimated functions. To demonstrate the strength and challenges, we estimated the DNNs using a stated preference survey from Singapore and a revealed preference data from London, extracted the full list of economic information from the DNNs, and compared them with those from the DCMs. We found that the economic information either aggregated over trainings or population is more reliable than the disaggregate information of the individual observations or trainings, and that larger sample size, hyperparameter searching, model ensemble, and effective regularization can significantly improve the reliability of the economic information extracted from the DNNs. Future studies should investigate the requirement of sample size, better ensemble mechanisms, other regularizations and DNN architectures, better optimization algorithms, and robust DNN training methods to address DNNs three challenges to provide more reliable economic information for DNN-based choice models. [ABSTRACT FROM AUTHOR]

In response to severe traffic congestion and air pollution, Beijing introduced a car ownership restriction policy to curb growth in the number of private cars in the city. However, Beijing residents can still purchase and register their cars in neighboring cities and this "leakage" may substantially reduce the policy's effectiveness. Using city-level data collected from the CEIC China Premium Database, we aim to quantify the spill-over effect: the impact of Beijing's policy on the growth of private car registrations in neighboring cities. We first deploy a synthetic control method to create a weighted combination of non-treated cities for each treated city. We then employ a difference-in-differences approach to estimate the policy leakage. Our models suggest that the policy resulted in additional 443,000 cars sold in the neighboring cities (within 500 km of Beijing) from 2011 to 2013, compared to if the policy had not been implemented. 35–40% of the car growth reduction stipulated by the policy simply spilled over to neighboring cities. The significance of the policy leakage necessitates positioning Beijing's urban transportation in a broader context and executing regional collaboration. [ABSTRACT FROM AUTHOR]

The bus control problem that combines holding and stop-skipping strategies is formulated as a multi-agent reinforcement learning (MARL) problem. Traditional MARL methods, designed for settings with joint action-taking, are incompatible with the asynchronous nature of at-stop control tasks. On the other hand, using a fully decentralized approach leads to environment non-stationarity, since the state transition of an individual agent may be distorted by the actions of other agents. To address it, we propose a design of the state and reward function that increases the observability of the impact of agents' actions during training. An event-based mesoscopic simulation model is built to train the agents. We evaluate the proposed approach in a case study with a complex route from the Chicago transit network. The proposed method is compared to a standard headway-based control and a policy trained with MARL but with no cooperative learning. The results show that the proposed method not only improves level of service but it is also more robust towards uncertainties in operations such as travel times and operator compliance with the recommended action. • A novel framework is proposed for joint holding and stop-skipping controls. • The framework is decentralized and considers nearby agents for evaluation. • Simulation experiments are conducted for a busy route operation in Chicago. • Metrics relevant to riders and agencies show the superiority of the proposed method. • The method shows robustness to reduced operator compliance. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) have been increasingly applied in travel demand modeling because of their automatic feature learning, high predictive performance, and economic interpretability. Nevertheless, DNNs frequently present behaviorally irregular patterns, significantly limiting their practical potentials and theoretical validity in travel behavior modeling. This study proposes strong and weak behavioral regularities as novel metrics to evaluate the monotonicity of individual demand functions (known as the "law of demand"), and further designs a constrained optimization framework with six gradient regularizers to enhance DNNs' behavioral regularity. The empirical benefits of this framework are illustrated by applying these regularizers to travel survey data from Chicago and London, which enables us to examine the trade-off between predictive power and behavioral regularity for large versus small sample scenarios and in-domain versus out-of-domain generalizations. The results demonstrate that, unlike models with strong behavioral foundations such as the multinomial logit, the benchmark DNNs cannot guarantee behavioral regularity. However, after applying gradient regularization, we increase DNNs' behavioral regularity by around 6 percentage points while retaining their relatively high predictive power. In the small sample scenario, gradient regularization is more effective than in the large sample scenario, simultaneously improving behavioral regularity by about 20 percentage points and log-likelihood by around 1.7%. Compared with the in-domain generalization of DNNs, gradient regularization works more effectively in out-of-domain generalization: it drastically improves the behavioral regularity of poorly performing benchmark DNNs by around 65 percentage points, highlighting the criticality of behavioral regularization for improving model transferability and applications in forecasting. Moreover, the proposed optimization framework is applicable to other neural network–based choice models such as TasteNets. Future studies could use behavioral regularity as a metric along with log-likelihood, prediction accuracy, and F 1 score when evaluating travel demand models, and investigate other methods to further enhance behavioral regularity when adopting complex machine learning models. • Designing behavioral regularity metrics using the law of demand. • Using behavioral regularity to evaluate deep learning in choice analysis. • Instilling behavioral regularity into deep learning through gradient regularization. • Empirically, gradient regularization improves behavioral regularity of deep learning. • Gradient regularization also enhances prediction in small samples. [ABSTRACT FROM AUTHOR]

As Transportation Network Companies (TNCs) have expanded their role in U.S. cities recently, their services (i.e. ridehailing) have been subject to scrutiny for displacing public transit (PT) ridership. Previous studies have attempted to classify the relationship between transit and TNCs, though analysis has been limited by a lack of granular TNC trip records, or has been conducted at aggregated scales. This study seeks to understand the TNC-PT relationship in Chicago at a spatially and temporally granular level by analyzing detailed individual trip records. An analysis framework is developed which enables TNC trips to be classified according to their potential relationship with transit: complementary (providing access to/from transit), substitutive (replacing a transit alternative), or independent (not desirably completable by transit). This framework is applied to both regular operating conditions and to early stages of the COVID-19 pandemic, to identify the TNC-PT relationship in these two contexts. We find that complementary TNC trips make up a small fraction of trips taken (approximately 2%), while potential independent trips represent 48% to 53% and potential substitution trips represent 45% to 50%. The percentage of substitution trips drops substantially following COVID-19 shutdowns (to around 14%). This may be attributed to a reduction in work-based TNC trips from Chicago's north side, indicated by changes in spatial distributions and flattening of trips occurring during peak hours. Furthermore, using spatial regression, we find that an increased tendency of TNC trips to substitute transit is related to a lower proportion of elderly people, greater proportion of peak-period TNC travel, greater transit network availability, a higher percentage of white population, and increased crime rates. Our findings identify spatial and temporal trends in the tendency to use TNC services in place of public transit, and thus have potential policy implications for transit management, such as spatially targeted service improvements and safety measures to reduce the possibility of public transit being substituted by TNC services. • We studied TNC-PT relationship at a spatially and temporally granular level. • TNC-PT relationship before and during the COVID-19 shutdown is examined. • Only approximately 2% of TNC trips complement public transit. • 45% to 50% of TNC trips substitute transit and this percentage drops during COVID-19 shutdowns. • Crime rate positively correlate with TNC-PT substitution both before and during COVID-19 shutdowns. [ABSTRACT FROM AUTHOR]

Shipment insurance has been widely used in express logistics and airline transportation. If a customer purchases a shipment insurance service and a disruption occurs, he could be compensated based on the declared value. This paper studies two contracts and investigates how shipment insurance premium affects the ex ante declared value and the ex post compensation. The customer purchases shipment insurance only when the cargo value is relatively high and his declared value does not exceed the actual cargo value. The optimal insurance premiums for both contracts and contract preference are obtained. Finally, we investigate the impact of effort towards transport process improvement. [ABSTRACT FROM AUTHOR]

A novel Cu 2 O/{101}TiO 2 nanosheet (Cu 2 O/{101}TNS) array film was prepared by the electrodeposition of Cu 2 O on the {101} facets of anatase nanosheet arrays by varying electrodeposition potential. The crystal structure, morphology, elemental chemical states, optical properties, photoelectrochemical properties, and stability of Cu 2 O/{101}TNS array films were investigated in detail. For TNS, due to different band structures and band edge positions between {001} and {101} facets, Cu 2 O/{101} facets of TiO 2 have higher band offset value which supply a larger driving force to increase the transport efficiency of carriers. Besides, owing to the directional flow of photo-generated electrons from {001} to {101} facets, the electrodeposition of Cu 2 O on the {101} facets of TNS will shorten the route length that the electrons must travel, thus reduce recombination of photo-generated electron-hole pairs. In addition, as the applied negative potential is high enough, a part of Cu + is reduced to Cu, which is beneficial for the photoexcited electrons transfer from CB of Cu 2 O to that of TiO 2 . The enhanced photoelectrochemical properties of Cu 2 O/{101}TNS array films can be attributed to the cocontributions of different band edge positions between {001} and {101} facets, Cu 2 O-Cu-TiO 2 ternary components and vertically aligned single-crystal TiO 2 nanosheet structure. [ABSTRACT FROM AUTHOR]