Actual Route Research Articles

Enhancing last-mile delivery: a hybrid approach with machine learning techniques that captures drivers' knowledge

ABSTRACT The rise of e-commerce has transformed last-mile delivery, with companies’ prioritising faster, more flexible options and implementing innovations such as route optimisation. Efficient last-mile delivery is now critical to customer satisfaction and business success. This work aims to bridge the gap between planned and actual delivery routes, a challenge highlighted by the 2021 Amazon Last-Mile Routing Research Challenge. The solution uses a sophisticated hybrid approach, combining machine learning algorithms with automated hyperparameter optimisation. Instead of focusing on individual stops, it predicts sequences of zones. The process involves data pre-processing, a Prediction by a Partial Matching algorithm to identify optimal zone combinations, a Rollout Algorithm to compute zone sequences for unexplored routes, and a Lin-Kernighan-Helsgaun solver for zone-to-zone routing. These steps are seamlessly integrated into a repeatable pipeline that automates hyperparameter fine-tuning. The results obtained indicate a robust solution capable of producing high-quality predictions.

Safety and bicycle route choice – To what extent does accident risk and perceived safety influence bicycle route choice?

Safety is a major issue in bicycle traffic and understanding safety-related factors that influence bicycle route choice is crucial for improving safety and promoting sustainable transportation. However, there is a notable lack of research on the impact of accident risk and perceived safety on cyclists' route choice. This study addresses this gap by investigating whether accident-prone areas or perceived insecure locations affect actual route choice decisions. The contribution explores this relationship by leveraging an extensive dataset comprising approximately 4,000 trips from around 170 participants, alongside additional data on infrastructure, operations, exposure, accidents, and mobility diaries on critical incidents. The findings broadly confirm the results from other studies regarding the influence of route characteristics (e.g., existence of cycling infrastructure, traffic volumes of motorized traffic, or presence of signal-controlled intersections). Moreover, the study reveals that a high accident risk along a route does have a slight negative influence on route choice. Surprisingly, perceived safety does not significantly influence route choice of cyclists.

Travel route recommendation with a trajectory learning model

This study addresses a critical issue in location-based services: travel route recommendation. It leverages historical trajectory data to predict the actual route on a road network from a starting point to a destination, given a specific departure time. However, capturing the latent patterns in complex trajectory data for accurate route planning presents a significant challenge. Existing route recommendation methods commonly face two major problems: first, inadequate integration of multi-source data, which fails to fully consider the potential factors affecting route choice; and second, limited capability to capture road network characteristics, which restricts the effective application of node features and negatively impacts recommendation accuracy. To address these issues, this research introduces a Trajectory Learning Model for Route Recommendation (TLMR) based on deep learning techniques. TLMR enhances the understanding of user route choice behavior in complex environments by integrating multi-source data. Moreover, by incorporating road network features, TLMR more effectively captures and utilizes the structural and dynamic information of the road network. Specifically, TLMR first employs a Position-aware Graph Neural Network to learn features of intersections from the road network, incorporating context features like weather and traffic conditions. Then, it integrates this information through neural networks to predict the next intersection. Finally, a beam search algorithm is applied to generate and recommend multiple candidate routes. Extensive experiments on four large real-world datasets demonstrate that TLMR outperforms existing methods in four key performance metrics. These results prove the effectiveness and superiority of TLMR in route recommendation.

Modeling Civil Aviation Emissions with Actual Flight Trajectories and Enhanced Aircraft Performance Model

Aviation emissions are continuously increasing along with the rapid development of air transportation, and results in the deterioration in regional air quality and the global climate. Accurate emission estimation is of great importance for relevant policies promotion and the sustainable development of the environment. Previous studies focused on the total emissions of a flight and lacked high precision in both spatial and temporal resolutions, especially aviation activities near ground. In this research, we propose an open-sourced emission calculation framework based on actual flight trajectories (TrajEmission), which calculates both the ground and airborne emissions simultaneously according to the configuration parameters, trajectory characteristics, and ambient conditions. We compare the emission results with five emission inventory methods. The results indicate that pollutant (nitrogen oxides, carbon monoxide, and unburned hydrocarbons) emissions in the landing and takeoff (LTO) cycle might usually be underestimated due to a lack of trajectory-based methods. In addition, in the overall results, the method based on the great circle route leads to an overestimation of 56.8% of pollutant emissions compared to the method based on actual routes. We also investigate the extent to which other factors could influence the emission results. To summarize, the TrajEmission framework can build inventories for the whole process of flight movements with high spatial–temporal resolutions and provide solid data support for environmental science and other related fields.

The carbon footprint of vegetable imports into Aruba: A closer look at sea and air transport

This study aimed to give insights into low-carbon vegetable import strategies for Aruba, a Dutch Caribbean island. Our selected products were potatoes, lettuce, onions, tomatoes, and green beans. The products originated from 13 different countries, and 25 product-country combinations were identified. The system boundaries were from the farm until arrival at the supermarket. We identified actual maritime transport routes, and calculated greenhouse gas (GHG) emissions of passenger aircrafts flying from Amsterdam to Aruba. Vegetables imported by air had significantly higher GHG emissions (4.2–8.3 kg CO2eq per kg) than products imported by sea (0.4–2.3 kg CO2eq per kg). GHG emissions of road transport generally contributed more than those of other life cycle stages, except when products showed a high contribution of agriculture. Although sea transport was calculated with much detail, it usually did not contribute much to the GHG emissions. We recommend Life Cycle Practitioners to consider aircraft characteristics when calculating GHG emissions of air transport, and to include the weight of the 80 kg AKE container, used for cooled airfreight, when allocating impacts between passengers and freight. For this case study, GHG emissions of specific passenger aircrafts always resulted in lower GHG emissions compared to generic calculations.

Searching Method for Three-Dimensional Puncture Route to Support Computed Tomography-Guided Percutaneous Puncture.

In CT-guided percutaneous punctures-an image-guided puncture method using CT images-physicians treat targets such as lung tumors, liver tumors, renal tumors, and intervertebral abscesses by inserting a puncture needle into the body from the exterior while viewing images. By recognizing two-dimensional CT images prior to a procedure, a physician determines the least invasive puncture route for the patient. Therefore, the candidate puncture route is limited to a two-dimensional region along the cross section of the human body. In this paper, we aim to construct a three-dimensional puncture space based on multiple two-dimensional CT images to search for a safer and shorter puncture route for a given patient. If all puncture routes starting from a target in the three-dimensional space were examined from all directions (the brute-force method), the processing time to derive the puncture route would be very long. We propose a more efficient method for three-dimensional puncture route selection in CT-guided percutaneous punctures. The proposed method extends the ray-tracing method, which quickly derives a line segment from a given start point to an end point on a two-dimensional plane, and applies it to three-dimensional space. During actual puncture route selection, a physician can use CT images to derive a three-dimensional puncture route that is safe for the patient and minimizes the puncture time. The main novelty is that we propose a method for deriving a three-dimensional puncture route within the allowed time in an actual puncture. The main goal is for physicians to select the puncture route they will use in the actual surgery from among the multiple three-dimensional puncture route candidates derived using the proposed method. The proposed method derives a three-dimensional puncture route within the allowed time in an actual puncture. Physicians can use the proposed method to derive a new puncture route, reducing the burden on patients and improving physician skills. In the evaluation results of a computer simulation, for a 3D CT image created by combining 170 two-dimensional CT images, the processing time for deriving the puncture route using the proposed method was approximately 59.4 s. The shortest length of the puncture route from the starting point to the target was between 20 mm and 22 mm. The search time for a three-dimensional human body consisting of 15 CT images was 4.77 s for the proposed method and 2599.0 s for a brute-force method. In a questionnaire, physicians who actually perform puncture treatments evaluated the candidate puncture routes derived by the proposed method. We confirmed that physicians could actually use these candidates as a puncture route.

Walking to School: What Streets Do Children Prefer?

Active school travel provides children with a daily opportunity to engage actively with their local urban environments. Despite widespread recognition that understanding the underlying factors of children’s navigation choices is crucial for developing effective environmental interventions to promote active school travel, there is limited evidence on children’s experiences regarding their school journeys. This is due in part to the fact that most studies rely on GIS-calculated routes which may not adequately represent children’s actual home-school journeys, and hence actual experiences. This study aims to identify specific environmental attributes influencing children’s navigation choices based on children’s (9–10 year olds) actual walking routes to school in Newcastle upon Tyne, UK. 45 pairs of selected and avoided streets were compared using a range of urban form (e.g., street connectivity measured through space syntax variables) and street design (e.g., footpath width) characteristics. Statistical analysis highlighted significant design attributes as potential determinants of navigation selections. In-depth street-level observations provided insight into the urban character of these street pairs, identifying the environmental qualities that could offer opportunities for active and safe commuting among children. This study contributes to the literature by broadening our understanding of the environmental attributes that may promote active school travel. Our findings, based on children’s actual experiences, may also inform urban planners and designers on designing inclusive child-friendly cities.

Perancangan Rute Optimal Distribusi Kargo Bahan Bakar Minyak dengan Metode Brute - Force Untuk Meminimasi Biaya Konsumsi Bahan Bakar Minyak Perusahaan Pelayaran PT XYZ

TSP of fuel oil distribution using tankers, the projected route search often does not contain the most minimal mileage, which can lead to adverse results such as significant additional costs. This research analyzes four different routes using several iterative proposals to reduce travel distance, time, and fuel costs compared to actual conditions. Problem solving is carried out using the Brute-Force method which enumerates all permutation results. The analysis results show that all proposed iterations were able to provide significant reductions in all three aspects. The four research routes achieved optimization results of 21,7% compared to the actual route modeling. This optimization proves that the iterative methods applied can substantially improve operational efficiency. Therefore, the determination of the tanker route must consider all the possibilities required in determining the route. The optimal route created is expected to have a new model with the aim of minimizing fuel costs based on the smallest distance and time calculations.

Disjoint paths construction algorithm in the data center network DPCell

Abstract With the development of the fourth industrial revolution, the importance of data centers has significantly increased. Data centers are widely used in many fields due to their ability to provide efficient, secure, and reliable data storage and processing services. However, with the increasing amount of data, traditional data center networks (DCNs) are currently facing various challenges, prompting academia and industry to propose new DCN architectures. As a dual-port server-based DCN, DPCell has excellent scalability and bisection width, enabling it to meet the demands of large-scale data storage, processing, and computation in the digital revolution. In order to ensure the secure and reliable data communication in the DPCell, this paper designs a disjoint paths communication scheme based on the actual DCN routing requirements. This scheme constructs the optimal number of disjoint paths in DPCell, with a maximum path length of $2^{k}+3$, where $k$ represents the dimension of the DPCell. Furthermore, experiments have verified that the time complexity of this scheme is sublinear, making it more efficient than the current optimal maximum flow algorithm. To a certain extent, this scheme provides DPCell with the required high bandwidth, fault tolerance, and security for data communication.

INNOVATIVE BY-PASS SOLUTION FOR IRON GATES I TO RECONNECT THE HISTORICAL MIGRATION ROUTES OF WILD ANADROMOUS STURGEON SPECIES, IN ORDER TO IMPROVE THEIR CONSERVATION STATUS

The anadromous sturgeon species in the lower Danube are currently critically endangered and based on the studies of the INCDPM expert team and the 2022 report of IUCN-SSG (Sturgeon Specialist Group) it appears that the future prospects of these populations are unfavourable. This situation occurs mainly as a result of anthropogenic impacts, especially due to intensive fishing, poaching and hydrotechnical constructions (e.g.: dams, immersed obstacles such as the bottom sill on Bala Branch). The dams interrupted the historical spawning migration routes of these species and the bottom sill can interrupt them in the future. In this context, INCDPM has elaborated an innovative by pass solution for the Iron Gates I, based on the longterm monitoring of ultrasonically tagged sturgeons from the Black Sea to the Iron Gates. This article presents the methodology developed by INCDPM in order to ensure the implementation of the innovative by-pass solution with minimal risk of unseccesful restoration of the historical migration routes. The study aims to present the specific migration conditions and the actual routes, which have been determined by over a decade of in-situ monitoring of ultrasonically tagged specimens. Additionally, simulations of upstream migration will be conducted and GIS maps will be developed to present historical migratory routes, with consideration given to hydrodynamic conditions and climate change scenarios.

Where should we go - Estimating travel times for modelling accessibility to 24-hour emergency departments in Canada

Estimating travel time to 24-hour emergency services is an important component to modelling accessibility of health services, particularly for rural areas. However, methods used to estimate travel time vary significantly, are not representative of the residential population, and are not openly validated. This makes the assessment of travel-based accessibility metrics between studies incomparable. To address this issue and develop a standardized measurement of emergency service access, this study utilized small geographic units (Dissemination Areas – DA) and geographical boundaries representative of municipal equivalents (Census Subdivision – CSD). Estimated travel times between the centroid of an inhabited DA to each 24-hr emergency department was computed with population-weighted travel times generated for each CSD. This dataset provides a nationally consistent measurement of proximity to emergency services accounting for travel pathing and population distribution. This methodology can be extended to generate estimated shortest travel routes for other healthcare resources or develop actual travel routes based on individuals’ experiences with the healthcare system.

Estimating the synthetic accessibility of molecules with building block and reaction-aware SAScore

Synthetic accessibility prediction is a task to estimate how easily a given molecule might be synthesizable in the laboratory, playing a crucial role in computer-aided molecular design. Although synthesis planning programs can determine synthesis routes, their slow processing times make them impractical for large-scale molecule screening. On the other hand, existing rapid synthesis accessibility estimation methods offer speed but typically lack integration with actual synthesis routes and building block information. In this work, we introduce BR-SAScore, an enhanced version of SAScore that integrates the available building block information (B) and reaction knowledge (R) from synthesis planning programs into the scoring process. In particular, we differentiate fragments inherent in building blocks and fragments to be derived from synthesis (reactions) when scoring synthetic accessibility. Compared to existing methods, our experimental findings demonstrate that BR-SAScore offers more accurate and precise identification of a molecule's synthetic accessibility by the synthesis planning program with a fast calculation time. Moreover, we illustrate how BR-SAScore provides chemically interpretable results, aligning with the capability of the synthesis planning program embedded with the same reaction knowledge and available building blocks.Scientific contributionWe introduce BR-SAScore, an extension of SAScore, to estimate the synthetic accessibility of molecules by leveraging known building-block and reactivity information. In our experiments, BR-SAScore shows superior prediction performance on predicting molecule synthetic accessibility compared to previous methods, including SAScore and deep-learning models, while requiring significantly less computation time. In addition, we show that BR-SAScore is able to precisely identify the chemical fragment contributing to the synthetic infeasibility, holding great potential for future molecule synthesizability optimization.

Programmable metasurfaces with high angular stability for arbitrarily-polarized obliquely-incident waves

Programmable metasurfaces show enormous potential in real-time electromagnetic (EM) manipulation and their stable responses to variable EM waves including incident angle and polarization changes are crucial for related applications. However, the demonstrated programmable metasurfaces are commonly considered in normal-incident waves with fixed polarization; how to achieve stable performance in more realistic scenarios of wide-angle and full-polarized wave incidences is still a challenge. Here, we propose and realize a wide-angle and full-polarization programmable metasurface, which can generate stabilized amplitude and phase responses at both transverse electric (TE) and transverse magnetic (TM) oblique incidences. These are achieved by introducing metallic walls around the metasurface element to reduce element coupling, which greatly improves the angular stability of its reflection responses. The mechanisms of angular insensitivity are analyzed comprehensively, and as a proof of concept, obliquely-incident beam steering, large-angle beam scanning and oblique vortex beam emitting are demonstrated on this programmable metasurface. Both the simulation and experimental results demonstrate that the programmable metasurface can operate well at both TE and TM wide-angle oblique incidences in the upper half space. Our work offers an actual route for improving the angular stability and polarization insensitivity of metasurfaces, which could push them one step closer towards more complicated applications.

Green commutes: Assessing the associations between green space exposure along GPS-track commuting routes and adults’ self-perceived stress

BackgroundHabitual commutes through green space may be associated with health promotion, including stress alleviation. However, few studies have assessed green space exposure during commuting with stress levels, and none have tracked commuters’ actual routes. AimTo assess 1) the association between commuting through green space and people's self-perceived stress, 2) whether the association is moderated by the transport mode, and 3) whether the association depends on differing green space operationalizations and buffer sizes. MethodsThis cross-sectional study used questionnaires and global positioning system data from Dutch adults aged 18–65 (N = 275). The Perceived Stress Scale was used to measure people's stress levels. Green space was measured by calculating the green space percentage from land use data (GREENLU), and by using normalized difference vegetation index (NDVI) obtained from Sentinel-2 (NDVISE) and Landsat-8 (NDVILS) satellite imagery. Buffers of 50, 100, and 250 m along commuting routes were applied to assess subjects’ environmental exposures. Associations were estimated using ordinary least squares regression models. ResultsCovariate-adjusted regressions showed that GREENLU was significantly but positively associated with stress levels, regardless of the buffer size. By contrast, the NDVI measures consistently showed null associations. We observed a positive association between GREENLU and stress levels among active commuters within the 250 m buffer in stratified analyses; however, associations were null for passive commuters across all green space measures and buffer sizes. ConclusionsOur findings suggest a counterintuitive positive association between increased exposure to green space during daily commutes and people's stress levels. These associations possibly may depend on the selected green space metric, the buffer sizes, and the commuting mode considered. The behavioral aspects of how people experience green environments, including commuting, may contribute to their impact on health.

Design Method of Cam Steering Mechanism Based on Path Fitting

In order to improve the accuracy of a solar-powered punch card car’s movement on a designated route and reduce positional deviations during its operation, a solar-powered punch card car with a single cam as the steering guidance mechanism was designed. The car adopts a three-wheel structure. The transmission mechanism, steering mechanism, driving mechanism, and regulating mechanism of the car were analyzed. The kinematics model of the car was established and the motion characteristics of the car were obtained. By analyzing the relationship between the steering angle of the car and the curvature radius of its travel route, the front wheel angle of the car at each position was calculated using MATLAB R2020a. This allowed us to establish the relationship between the front wheel angle and the displacement of the steering push rod, which was further converted into the theoretical contour line of the cam. Subsequently, the theoretical contour line of the cam was completed and envelope correction was performed. Finally, through mechanical analysis and experimental verification using a prototype, the results indicated that the single-cam steering guidance mechanism calculated using this fast path fitting method exhibited excellent mechanical performance and a smooth and accurate trajectory, and the traveling path of the theoretical cam contour curve was basically consistent with the actual trajectory route.

Traffic trajectory data analysis technology based on HMM model map matching algorithm.

The rapid growth of traffic trajectory data and the development of positioning technology have driven the demand for its analysis. However, in the current application scenarios, there are some problems such as the deviation between positioning data and real roads and low accuracy of existing trajectory data traffic prediction models. Therefore, a map matching algorithm based on hidden Markov models is proposed in this study. The algorithm starts from the global route, selects K nearest candidate paths, and identifies candidate points through the candidate paths. It uses changes in speed, angle, and other information to generate a state transition matrix that match trajectory points to the actual route. When processing trajectory data in the experiment, K = 5 is selected as the optimal value, the algorithm takes 51 ms and the accuracy is 95.3%. The algorithm performed well in a variety of road conditions, especially in parallel and mixed road sections, with an accuracy rate of more than 96%. Although the time loss of this algorithm is slightly increased compared with the traditional algorithm, its accuracy is stable. Under different road conditions, the accuracy of the algorithm is 98.3%, 97.5%, 94.8% and 96%, respectively. The accuracy of the algorithm based on traditional hidden Markov models is 95.9%, 95.7%, 95.4% and 94.6%, respectively. It can be seen that the accuracy of the algorithm designed has higher precision. The experiment proves that the map matching algorithms based on hidden Markov models is superior to other algorithms in terms of matching accuracy. This study makes the processing of traffic trajectory data more accurate.

Optimal Vehicle Scheduling and Charging Infrastructure Planning for Autonomous Modular Transit System

Prioritizing the development of public transport is an effective way to improve the sustainability of the transport system. In recent years, bus passenger flow has been declining in many cities. How to reform the operating mode of the public transportation system is an important issue that needs to be solved. An autonomous modular bus (AMB) is capable of physical coupling and uncoupling to flexibly adjust vehicle capacity as well as provide high-quality service under unbalanced passenger demand conditions. To promote AMB adoption and reduce the operating cost of the bus route, this paper presents a joint optimization method to simultaneously determine the AMB dispatching plan, charging plan, and charging infrastructure configuration scheme. Then, a mixed-integer programming model is formulated to minimize the operating costs of the bus route. A hybrid intelligent algorithm combining enumeration, cloning algorithm, and particle swarm optimization algorithm is designed to resolve the formulated model. Subsequently, an actual bus route is taken as an example to validate the proposed method. Results indicate that the developed method in this paper can reduce the operating costs and operational energy consumption of the route compared with the real route operating plan. Specifically, the reduction ratio of the former is 23.85%, and the reduction ratio of the latter is 5.92%. The results of this study validate the feasibility and advantages of autonomous modular transit service, contributing positively to the sustainable development of the urban public transportation system.

Optimizing Routes for Sustainability: A Comparative Analysis of Parcel Distribution Methods in South Jakarta

This study focuses on improving the efficiency of PT. XYZ’s logistics services by examining the optimization of parcel distribution routes in South Jakarta. It compares the Nearest Neighbor Method and the Clarke and Wright Savings Method to determine the optimal strategy for delivering 53 packages to 8 locations. The Nearest Neighbor Method generates two routes, one encompassing 32 packages across four locations and the other encompassing 21 packages across four locations. This method significantly reduces travel distance, saving 27,3 kilometers in comparison to actual courier routes. The Clarke and Wright Savings Method generates two routes, one covering 35 packages across six locations and the other covering 18 packages across two locations. Compared to the actual van courier route, this method reduces the distance by 24,8 kilometers. The Nearest Neighbor Method reduces travel distances by 27.3 kilometers compared to actual routes, surpassing the Clarke and Wright Savings Method by 2.5 kilometers. Reduced travel distances reduce fuel consumption and greenhouse gas emissions, making logistics operations more sustainable on an environmental and economic level. This study concludes that the Nearest Neighbor Method is the optimal strategy for improving parcel distribution efficiency in South Jakarta, thereby substantially reducing the travel distances of PT. XYZ’s logistics operations.

Recursive logit-based meta-inverse reinforcement learning for driver-preferred route planning

Driver-preferred route planning often evaluates the quality of a planned route based on how closely it is followed by the driver. Despite decades of research in this area, there still exist nonnegligible deviations from planned routes. Recently, with the prevalence of GPS data, Inverse Reinforcement Learning (IRL) has attracted much interest due to its ability to directly learn routing patterns from GPS trajectories. However, existing IRL methods are limited in that: (1) They rely on numerical approximations to calculate the expected state visitation frequencies (SVFs), which are inaccurate and time-consuming; and (2) They ignore the fact that the coverage of GPS trajectories is skewed toward popular road segments, causing difficulties in learning from sparsely covered ones. To overcome these challenges, we propose a recursive logit-based meta-IRL approach, where (1) We use the recursive logit model to capture drivers’ route choice behavior so that the expected SVFs can be analytically derived, which substantially reduces the computational efforts; and (2) We introduce meta-parameters and employ meta-learning techniques so that the learning on sparsely covered road segments can benefit from that on popular ones. When training our IRL model, we update the rewards of road segments with the expected SVFs by solving several systems of linear equations and update the meta-parameters through a two-level optimization structure to ensure its fast adaption and versatility. We validate our approach using real GPS data in Chengdu, China. Results show that our planned routes better match actual routes compared with state-of-the-art methods including the recursive logit model, Deep-IRL and Dij-IRL: the F1-Score increases by 4.17% with the introduction of the recursive logit model and further increases to 5.19% after meta-learning is employed. Moreover, we can reduce training time by over 95%.

Understanding the route choice preferences of private and dock-based public bike users using GPS data in Seoul, South Korea

Analyzing route preferences is crucial for comprehending how individuals choose their travel paths, assessing road facilities, and anticipating upcoming traffic conditions. This study aimed to explore the factors influencing route selections made by private and public bike users and compare their decisions during peak and off-peak hours. Global positioning system (GPS) traces of 12,106 private bike users and 74,397 public bike users from a single weekday in Seoul, South Korea, were used. A map-matching process aligned the GPS data with a comprehensive cycling network encompassing bikeways, sidewalks, neighborhood roads, and their street attributes such as urban, park, and river streets. Choice sets composed of the shortest and 10th alternative routes were created to compare various road features with actual routes. By employing a path-size logit model to address overlapping issues among the routes, the results showed that buffered bike lanes separated from car flows positively influenced the route selection of bicyclists. Bike boulevards and shared sidewalk bikeways negatively impacted route decisions made by public bike users, with a more pronounced effect during peak hours. Furthermore, private bike users showed a decreased preference for routes in residential areas during peak hours, while an increased preference during off-peak hours. Moreover, public bike users preferred selecting routes in the presence of shopping-stores or amenities as well as public transit stations, with these effects being more pronounced during peak hours. These findings are expected to contribute to capturing valuable insights into the route preferences of bicyclists for the improvement of bicycle infrastructures.