Railroad construction sites are high-risk environments where monitoring personnel safety is critical for preventing accidents and enhancing construction efficiency. Traditional manual monitoring and image processing methods exhibit deficiencies in real-time performance and accuracy. This paper proposes a railway worker detection method based on improved support vector machines (ISVM), while using non-local mean noise reduction and histogram equalisation pre-processing techniques to optimise image quality to improve detection efficiency and accuracy. Multiscale features are then extracted with Inception v3 and combined with principal component analysis (PCA) for dimensionality reduction. Finally, an SVM classification algorithm is employed for personnel detection. To process small sample categories, data enhancement techniques (e.g., random flip and rotation) and K-fold cross-validation are applied to optimize the model parameters. The experimental results demonstrate that the ISVM method significantly improves accuracy and real-time performance compared to traditional detection methods and single deep learning models. This method provides technical support for railroad construction safety monitoring and effectively addresses personnel detection tasks in complex construction environments.

Mixed non-motorized traffic is largely unaffected by motor vehicle congestion, offering high accessibility and convenience, and thus serving as a primary mode of “last-mile” transportation in urban areas. To advance stochastic capacity estimation methods and provide reliable assessments of non-motorized roadway capacity, this study proposes a stochastic capacity estimation model based on power spectral analysis. The model treats discrete traffic flow data as a time-series signal and employs a stochastic signal parameter model to fit stochastic traffic flow patterns. Initially, UAVs and video cameras are used to capture videos of mixed non-motorized traffic flow. The video data were processed with an image detection algorithm based on the YOLO convolutional neural network and a video tracking algorithm using the DeepSORT multi-target tracking model, extracting data on traffic flow, density, speed, and rider characteristics. Then, the autocorrelation and partial autocorrelation functions of the signal are employed to distinguish among four classical stochastic signal parameter models. The model parameters are optimized by minimizing the AIC information criterion to identify the model with optimal fit. The fitted parametric models are analyzed by transforming them from the time domain to the frequency domain, and the power spectrum estimation model is then calculated. The experimental results show that the stochastic capacity model yields a pure EV capacity of 2060–3297 bikes/(h·m) and a pure bicycle capacity of 1538–2460 bikes/(h·m). The density–flow model calculates a pure EV capacity of 2349–2897 bikes/(h·m) and a pure bicycle capacity of 1753–2173 bikes/(h·m). The minimal difference between these estimates validates the effectiveness of the proposed model. These findings hold practical significance in addressing urban road congestion.

Private vehicle overreliance in Klang Valley, Malaysia contributes to severe traffic congestion, air pollution, and carbon emissions, yet public transport modal shift remains far below national policy targets. This study extends the Theory of Planned Behaviour (TPB) by incorporating two external constructs, environmental concern (EC) and technology adoption (TA), to investigate the behavioural intention of 483 private vehicle users to switch to rail transit. Hypotheses were tested using partial least squares structural equation modelling (PLS-SEM). The results show that, with the exception of subjective norm ($\beta =0.064$, $p=0.180$), all constructs exert significant positive effects on behavioural intention. Technology adoption emerged as the strongest direct predictor ($\beta =0.417$), followed by perceived behaviour control ($\beta =0.340$). Environmental concern operated exclusively through indirect pathways, mediated by attitude and perceived behaviour control, and exhibited the largest effect size on the TPB components ($f^2$ = 0.380–0.449, large effect).

This paper focuses on predicting road passenger probability and optimizing taxi driving routes based on trajectory big data. By utilizing clustering algorithms to identify key passenger points, a method for calculating and predicting road passenger probability is proposed. This method calculates the passenger probability for each road segment during different time periods and uses a BiLSTM neural network for prediction. A passenger-seeking recommendation model is then constructed with the goal of maximizing passenger probability, and it is solved using the NSGA-II algorithm. Experiments are conducted on the Chengdu taxi trajectory dataset, using MSE as the metric for model prediction accuracy. The results show that the BiLSTM prediction model improves prediction accuracy by 9.67% compared to the BP neural network and by 6.45% compared to the LSTM neural network. The proposed taxi driver passenger-seeking route selection method increases the average passenger probability by 18.95% compared to common methods. The proposed passenger-seeking recommendation framework, which includes passenger probability prediction and route optimization, maximizes road passenger efficiency and holds significant academic and practical value.

Existing timetable optimisation models for urban rail transit predominantly adopt operator-oriented objectives with assumed passenger-related weights. This paper proposes a passenger satisfaction-oriented timetable optimisation framework in which satisfaction weights are empirically derived from confirmatory factor analysis and Cramér’s V analysis of survey data collected from 439 passengers on Nanning Rail Transit Line 1. Seven scheduling-related service attributes are formally expressed as functions of timetable decision variables, establishing a direct linkage between passenger perception and scheduling decisions. A multi-objective model minimises a weighted combination of passenger dissatisfaction, operational cost, and stranded passenger ratio, solved by a Passenger Satisfaction-oriented Adaptive Dispatch Heuristic (PS-ADH) integrating simulated annealing with a passenger flow simulation module. Case study results demonstrate simultaneous improvements of 3.78% in composite objective value, 3.25% in passenger dissatisfaction, and 3.25% in operational cost, with a 27.4% reduction in stranded passengers. The optimised strategy is selected consistently across all ten random initialisations (CV = 0.13%). Sensitivity analysis reveals a structural break at cost weight $\beta =0.4$, beyond which the optimal strategy shifts qualitatively toward cost minimisation at the expense of service quality. The framework provides a transferable methodology for integrating passenger perception data into rail transit scheduling for emerging urban rail systems.

Driven by low-carbon mandates, hybrid power vessels integrating diesel, battery, shore-power, and photovoltaic vessel (PV) systems are emerging as a key green shipping pathway. However, operation scheduling is essentially complex due to the integration between supply-side routing and load-side energy dispatch, which is compounded by multi-dimensional uncertainties in PV generation, port-grid loads, and feeder delays. To address this, we formulate a unified two-stage robust optimization model. The objective is to simultaneously minimize operating costs and enhance port-grid friendliness by coordinating on-board energy management and shore-power interactions. In detail, the first stage determines routing and sailing speeds, while the second stage allocates multi-source power under a worst-case budgeted polyhedral uncertainty set. A piecewise-linearization scheme handles the cubic speed–power relation, rendering a tractable mixed-integer linear programming problem. The problem is efficiently solved via a tailored Benders decomposition algorithm, utilizing a genetic-algorithm warm-start to substantially accelerate convergence. Validated on three real-world networks via 1000 Monte Carlo scenarios, the proposed model reduces mean operating costs by 20.2–23.5% and suppresses cost variance by over 60% compared to deterministic approaches.