ABSTRACT Reinforcement learning (RL) is a promising machine‐learning solution to traffic signal control problems, which have been extensively studied. However, variants of non‐linear, deep artificial neural network (ANN) function approximators (FAs) have been predominantly employed in previous studies proposing RL‐based controllers, leaving a significant interpretability issue due to their black‐box nature. In this work, the use of the linear FA for a value‐based RL agent in traffic signal control problems is investigated along with the least‐squares Q‐learning method, abbreviated as LSTDQ. The interpretable linear FA was found to be adequate for the RL agent to learn an optimal policy. This leads to the proposal to replace a non‐linear ANN FA with the linear FA counterpart, resolving the interpretability issue. Moreover, the LSTDQ learning method shows superior behaviour convergence compared to a gradient descent method. In a low‐intensity arrival pattern scenario, the control by the RL agent cuts about half of the average delay resulting from the pretimed control. Owing to the conciseness of the linear FA, a direct interpretation analysis of the converged linear‐FA parameters is presented. Lastly, two online relearning tests of the agents under non‐stationary arrivals are conducted to demonstrate the online performance of LSTDQ. In conclusion, the linear‐FA specification and the LSTDQ method are together proposed to be used for its control algorithm interpretability property, superior convergence quality, and lack of hyperparameters.
Boniphace Kutela, Frank Ngeni, Cuthbert Ruseruka
et al.
Over-speeding is a pivotal factor in fatal traffic crashes globally, necessitating robust speed management strategies to augment road safety. In 2021, the National Highway Traffic Safety Administration reported over 12 000 speed-related fatalities in the United States alone. Previous studies aggregated over-speeding tendencies; however, the extent of over-speeding has a significant implication on the crash outcome. This study delves into the prevalence and magnitude of over-speeding in various scenarios, utilizing data from traffic cameras in Edmonton, Canada, and employing a negative binomial statistical model for analysis. The model elucidates the significance and likelihood of over-speeding tendencies by incorporating temporal and built environment variables, i.e., year, month, number of lanes, dwelling unit types, school-related, and open green space. Study results indicated that the aggregation of the over-speeding data tends to underestimate the influence of various factors. Notably, the estimated impact of the posted speed limit for the disaggregated models is up to over two times that for the aggregated model. Further, the summer months exhibit a roughly 25% uptick in speed limit violations for aggregated models while about a 40% uptick in the speed limit violations for disaggregated approaches. Conversely, a discernible decline in over-speeding tendencies is observed with camera enforcement, showcasing a 25% reduction over four years. Built environment variables presented mixed results, with one-unit dwellings associated with a 12% increase in over-speeding, while proximity to schools indicated a 10% decrease. These pivotal findings provide policymakers and practitioners with valuable insights to formulate targeted interventions and countermeasures to curtail speed limit violations and bolster overall road safety conditions.
Countries worldwide are increasingly focused on addressing the imbalance between the supply and demand for EV charging infrastructure, with the community-shared charging post (CSCP) co-construction project emerging as a promising solution. The broad participation and investment support of the residents are the keys to the success of the CSCP co-construction project. This study, grounded in the theory of planned behavior (TPB) from social psychology, incorporated factors such as community identity, perceived green value, economic benefit, uncivil behaviors, and perceived risk to construct a structural model explaining community residents’ intention to invest in the CSCP co-construction project. This research confirmed that (1) 85.73% of respondents expressed strong recognition of the CSCP co-construction project, with a mean recognition score of 5.56 out of a possible 7; (2) an individual’s social-related perceptions, including the subjective norms and community identity are the strongest determinant of the intention to invest in the CSCP co-construction project; (3) the willingness to invest in CSCP co-construction project differs significantly between the EV group and the non-EV group. Economic benefit was significant only for the non-EV group, while uncivil behaviors were significant only for the EV group. These results provide valuable guidelines for governments and corporations that are promoting or pursuing sharing community for the residents.
Modern transportation systems face pressing challenges due to increasing demand, dynamic environments, and heterogeneous information integration. The rapid evolution of Large Language Models (LLMs) offers transformative potential to address these challenges. Extensive knowledge and high-level capabilities derived from pretraining evolve the default role of LLMs as text generators to become versatile, knowledge-driven task solvers for intelligent transportation systems. This survey first presents LLM4TR, a novel conceptual framework that systematically categorizes the roles of LLMs in transportation into four synergetic dimensions: information processors, knowledge encoders, component generators, and decision facilitators. Through a unified taxonomy, we systematically elucidate how LLMs bridge fragmented data pipelines, enhance predictive analytics, simulate human-like reasoning, and enable closed-loop interactions across sensing, learning, modeling, and managing tasks in transportation systems. For each role, our review spans diverse applications, from traffic prediction and autonomous driving to safety analytics and urban mobility optimization, highlighting how emergent capabilities of LLMs such as in-context learning and step-by-step reasoning can enhance the operation and management of transportation systems. We further curate practical guidance, including available resources and computational guidelines, to support real-world deployment. By identifying challenges in existing LLM-based solutions, this survey charts a roadmap for advancing LLM-driven transportation research, positioning LLMs as central actors in the next generation of cyber-physical-social mobility ecosystems. Online resources can be found in the project page: https://github.com/tongnie/awesome-llm4tr.
Large language models (LLMs) have rapidly gained popularity and are being embedded into professional applications due to their capabilities in generating human-like content. However, unquestioned reliance on their outputs and recommendations can be problematic as LLMs can reinforce societal biases and stereotypes. This study investigates how LLMs, specifically OpenAI's GPT-4 and Microsoft Copilot, can reinforce gender and racial stereotypes within the software engineering (SE) profession through both textual and graphical outputs. We used each LLM to generate 300 profiles, consisting of 100 gender-based and 50 gender-neutral profiles, for a recruitment scenario in SE roles. Recommendations were generated for each profile and evaluated against the job requirements for four distinct SE positions. Each LLM was asked to select the top 5 candidates and subsequently the best candidate for each role. Each LLM was also asked to generate images for the top 5 candidates, providing a dataset for analysing potential biases in both text-based selections and visual representations. Our analysis reveals that both models preferred male and Caucasian profiles, particularly for senior roles, and favoured images featuring traits such as lighter skin tones, slimmer body types, and younger appearances. These findings highlight underlying societal biases influence the outputs of LLMs, contributing to narrow, exclusionary stereotypes that can further limit diversity and perpetuate inequities in the SE field. As LLMs are increasingly adopted within SE research and professional practices, awareness of these biases is crucial to prevent the reinforcement of discriminatory norms and to ensure that AI tools are leveraged to promote an inclusive and equitable engineering culture rather than hinder it.
In recent years, open-pit mining has seen significant advancement, the cooperative operation of various specialized machinery substantially enhancing the efficiency of mineral extraction. However, the harsh environment and complex conditions in open-pit mines present substantial challenges for the implementation of autonomous transportation systems. This research introduces a novel paradigm that integrates Scenario Engineering (SE) with autonomous transportation systems to significantly improve the trustworthiness, robustness, and efficiency in open-pit mines by incorporating the four key components of SE, including Scenario Feature Extractor, Intelligence and Index, Calibration and Certification, and Verification and Validation. This paradigm has been validated in two famous open-pit mines, the experiment results demonstrate marked improvements in robustness, trustworthiness, and efficiency. By enhancing the capacity, scalability, and diversity of autonomous transportation, this paradigm fosters the integration of SE and parallel driving and finally propels the achievement of the ‘6S’ objectives.
In this paper, we explore the capabilities of state-of-the-art large language models (LLMs) such as GPT-4, GPT-4o, Claude 3.5 Sonnet, Claude 3 Opus, Gemini 1.5 Pro, Llama 3, and Llama 3.1 in solving some selected undergraduate-level transportation engineering problems. We introduce TransportBench, a benchmark dataset that includes a sample of transportation engineering problems on a wide range of subjects in the context of planning, design, management, and control of transportation systems. This dataset is used by human experts to evaluate the capabilities of various commercial and open-sourced LLMs, especially their accuracy, consistency, and reasoning behaviors, in solving transportation engineering problems. Our comprehensive analysis uncovers the unique strengths and limitations of each LLM, e.g. our analysis shows the impressive accuracy and some unexpected inconsistent behaviors of Claude 3.5 Sonnet in solving TransportBench problems. Our study marks a thrilling first step toward harnessing artificial general intelligence for complex transportation challenges.
Plant natural products (PNPs) exhibit a wide range of biological activities and have essential applications in various fields such as medicine, agriculture, and flavors. Given their natural limitations, the production of high-value PNPs using microbial cell factories has become an effective alternative in recent years. However, host metabolic burden caused by its massive accumulation has become one of the main challenges for efficient PNP production. Therefore, it is necessary to strengthen the transmembrane transport process of PNPs. This review introduces the discovery and mining of PNP transporters to directly mediate PNP transmembrane transportation both intracellularly and extracellularly. In addition to transporter engineering, this review also summarizes several auxiliary strategies (such as small molecules, environmental changes, and vesicles assisted transport) for strengthening PNP transportation. Finally, this review is concluded with the applications and future perspectives of transportation engineering in the construction and optimization of PNP microbial cell factories.
The increasing impact of the greenhouse effect on ecosystems is prompting transportation agencies to seek methods for reducing CO2 emissions during pavement construction and maintenance. Additionally, the laboratory mix design process, which involves selecting aggregate gradation and binder content, is time-consuming and labor-intensive. To accelerate the traditional mix design procedure, this study presented a mix design procedure that can automatically determine gradation and binder content based on machine learning (ML) and a meta-heuristic algorithm. Specifically, ML approaches were employed to model the relationship between volumetric properties (mixture bulk specific gravity (Gmb) and air void (VV)) and both mixture component properties and mixture proportion, based on a dataset collected from literature with 660 mixture designs. Integrated with the prediction of ML models and the modified multi-objective grey wolf optimization (MOGWO) algorithm, an automatic asphalt mix design was proposed to pursue three goals, including VV, cost, and CO2 emission. The results indicated that least squares support vector regression (LSSVR) and eXtreme gradient boosting (XGBoost) achieved the highest prediction accuracies (correlation coefficient: 0.92 for VV and 0.96 for Gmb). The MOGWO algorithm successfully found the 26 optimal mix designs for the case of VV vs. cost vs. CO2 emission. Compared to the traditional laboratory design, the optimal mixture with VV of 4% achieves a cost saving of 2.46% and a reduction of 4.03% in carbon emission. The volumetric properties of the mixtures output by the approach also align closely with values measured in a laboratory.
Purpose. The paper aims to highlight and substantiate the need to find rational design schemes for lattice tube concrete bridges with a ride on top based on the analysis of recent research and regulatory documents. Methodology. The current scientific research is analyzed to determine the current state of development of pipe concrete lattice structures. Methods for improving structures are presented. Combinations of filling the grating elements with concrete, variants of cross-sections of the grating elements, their advantages and disadvantages are analyzed. The state of building codes of Ukraine and other countries is considered in order to determine possible options for the design of pipe concrete bridge structures. Due to the lack of detailed research on this issue, the feasibility of implementing optimization studies for these structures and the steps necessary for this are determined. Findings. The optimization of pipe-concrete bridge structures is a relevant area of research, but it requires a multicomponent approach and the use of modern computer facilities. The method of linear optimization is proposed and its general steps for finding economic models are determined. It was found that the base of Ukrainian SCSs in the field of pipe and concrete structures is limited, but can be expanded by using European standards and other international regulations. Originality. The necessity of global development and improvement of pipe concrete gratings of bridge spans is highlighted. Attention is focused on the advantages of this area, which contributes to decision-making at the stage of selecting the type of bridge and detailed design of pipe-concrete lattice bridges. A methodology for finding the optimal grids is proposed, which can integrate existing methods of structural improvement and the requirements of regulatory documents. Practical value. The results of the study can be used to improve the design of pipe concrete bridges at the design stage. Optimization of gratings can help to increase the efficiency of construction and reliability of this type of bridge structure.
Abstract Crowdfunding has become important in increasing financial support for the development of green technologies. Self-disclosed information significantly affects supporters’ decisions and is important for the success of green project funding. However, current studies still lack investigations into the impact of information disclosure on green crowdfunding performance. This research aims to fill this knowledge gap by exploring eight information disclosure-relevant factors in green crowdfunding performance. Applying machine learning techniques (e.g., Natural Language Processing and Computer Vision) and logistic regression, this study investigates 720 green crowdfunding campaigns on GoFundMe and empirically finds that the duration, length of campaign introductions, and length of the title influence fundraising outcomes. However, no evidence supports the impact of goal size, emotion of campaign introduction, or image content on funding success. This study clarifies the information disclosure-related data that green crowdfunding campaigns should consider and provides founders with a constructive guide to smoothly raise money for a green crowdfunding campaign. This study also contributes to data processing methods by providing future studies with an approach for transferring unstructured data to structured data.
The ability of Generative AI (GAI) technology to automatically check, synthesize and modify software engineering artifacts promises to revolutionize all aspects of software engineering. Using GAI for software engineering tasks is consequently one of the most rapidly expanding fields of software engineering research, with over a hundred LLM-based code models having been published since 2021. However, the overwhelming majority of existing code models share a major weakness - they are exclusively trained on the syntactic facet of software, significantly lowering their trustworthiness in tasks dependent on software semantics. To address this problem, a new class of "Morescient" GAI is needed that is "aware" of (i.e., trained on) both the semantic and static facets of software. This, in turn, will require a new generation of software observation platforms capable of generating large quantities of execution observations in a structured and readily analyzable way. In this paper, we present a vision and roadmap for how such "Morescient" GAI models can be engineered, evolved and disseminated according to the principles of open science.
Roberto Casadei, Gianluca Aguzzi, Giorgio Audrito
et al.
Today's distributed and pervasive computing addresses large-scale cyber-physical ecosystems, characterised by dense and large networks of devices capable of computation, communication and interaction with the environment and people. While most research focusses on treating these systems as "composites" (i.e., heterogeneous functional complexes), recent developments in fields such as self-organising systems and swarm robotics have opened up a complementary perspective: treating systems as "collectives" (i.e., uniform, collaborative, and self-organising groups of entities). This article explores the motivations, state of the art, and implications of this "collective computing paradigm" in software engineering, discusses its peculiar challenges, and outlines a path for future research, touching on aspects such as macroprogramming, collective intelligence, self-adaptive middleware, learning, synthesis, and experimentation of collective behaviour.
A paradigm shift is underway in Software Engineering, with AI systems such as LLMs playing an increasingly important role in boosting software development productivity. This trend is anticipated to persist. In the next years, we expect a growing symbiotic partnership between human software developers and AI. The Software Engineering research community cannot afford to overlook this trend; we must address the key research challenges posed by the integration of AI into the software development process. In this paper, we present our vision of the future of software development in an AI-driven world and explore the key challenges that our research community should address to realize this vision.
With the proliferation of electric vehicles (EVs), the transportation network and power grid become increasingly interdependent and coupled via charging stations. The concomitant growth in charging demand has posed challenges for both networks, highlighting the importance of charging coordination. Existing literature largely overlooks the interactions between power grid security and traffic efficiency. In view of this, we study the en-route charging station (CS) recommendation problem for EVs in dynamically coupled transportation-power systems. The system-level objective is to maximize the overall traffic efficiency while ensuring the safety of the power grid. This problem is for the first time formulated as a constrained Markov decision process (CMDP), and an online prediction-assisted safe reinforcement learning (OP-SRL) method is proposed to learn the optimal and secure policy by extending the PPO method. To be specific, we mainly address two challenges. First, the constrained optimization problem is converted into an equivalent unconstrained optimization problem by applying the Lagrangian method. Second, to account for the uncertain long-time delay between performing CS recommendation and commencing charging, we put forward an online sequence-to-sequence (Seq2Seq) predictor for state augmentation to guide the agent in making forward-thinking decisions. Finally, we conduct comprehensive experimental studies based on the Nguyen-Dupuis network and a large-scale real-world road network, coupled with IEEE 33-bus and IEEE 69-bus distribution systems, respectively. Results demonstrate that the proposed method outperforms baselines in terms of road network efficiency, power grid safety, and EV user satisfaction. The case study on the real-world network also illustrates the applicability in the practical context.
A central challenge for ensuring the success of software projects is to assure the convergence of developers' and users' views. While the availability of large amounts of user data from social media, app store reviews, and support channels bears many benefits, it still remains unclear how software development teams can effectively use this data. We present an LLM-powered approach called DeeperMatcher that helps agile teams use crowd-based requirements engineering (CrowdRE) in their issue and task management. We are currently implementing a command-line tool that enables developers to match issues with relevant user reviews. We validated our approach on an existing English dataset from a well-known open-source project. Additionally, to check how well DeeperMatcher works for other languages, we conducted a single-case mechanism experiment alongside developers of a local project that has issues and user feedback in Brazilian Portuguese. Our preliminary analysis indicates that the accuracy of our approach is highly dependent on the text embedding method used. We discuss further refinements needed for reliable crowd-based requirements engineering with multilingual support.