Clustering of public transport operation using K-means

Kadir, Rabiah Abdul; Shima, Yasuki; Sulaiman, Riza; Ali, Fathelalem F.

doi:10.1109/icbda.2018.8367721

Cited by 11 publications

(9 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To start, the total number of clusters is equal to the total observations. This clustering method continues to combine the most similar clusters into a new one ( 18 ). The output of using a hierarchical method is a set of stages where the grouped clusters and the distance between clusters are identified.…”

Section: Methodsmentioning

confidence: 99%

“…A dendrogram is used to visualize the results and to choose the best sets that should be clustered. The “most dominant consideration” is the distance calculated between clusters ( 18 ). This distance increases with the stages as grouping the clusters results in having more widespread elements within each newly updated cluster.…”

Section: Methodsmentioning

confidence: 99%

“…However, the campus shuttle scheduling problem does not seem to have received enough attention. Literature shows that some studies used the clustering method for finding homogenous time periods; they aggregated the periods to have several 1-h, or 30-min periods, and they did not consider effects from the class schedules (17)(18)(19). Also, they used the origin-destination demand, or the total volume separately as a variable for clustering.…”

Section: Literature Reviewmentioning

confidence: 99%

See 2 more Smart Citations

Cluster Based Methodology for Scheduling a University Shuttle System

Mahmoudzadeh

Wang

2020

Transportation Research Record

View full text Add to dashboard Cite

A university transit service owns and operates a shuttle fleet that serves thousands of students that reside on and around the campus within a fairly compact area as a special form of public transit. Usually, there is a central drop off point in the center of the campus. Each shuttle goes on a tour that covers a certain area where apartment complexes housing the students are located. The main goal of this study is to find a practical scheduling methodology for the university shuttle fleet considering students’ travel behaviors and class times. The 2018–2019 annual ridership data of Texas A&M University (TAMU) shuttle system is used to identify distinct travel patterns in weekday and weekend services. A clustering method is proposed to develop homogenous departure times within the current schedules respectively for on- and off-campus routes by taking the high student loads and graduate/undergraduate behaviors into account. A new scheduling method is developed to update the current shuttle departure times by leaning on a generous/highly frequent system usage. This easy to use method considers the dynamics of daily activities conveniently. Results of the developed new schedules for on-campus and off-campus routes show that the current university shuttle fleet could operate more efficiently during the time when class changes take place. This new methodology could potentially increase the efficiency of the system in relation to total shuttle runs by up to 25%.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Cluster Based Methodology for Scheduling a University Shuttle System

Mahmoudzadeh

Wang

2020

Transportation Research Record

View full text Add to dashboard Cite

show abstract

“…K -means clustering as a means of uncovering underlying patterns in unlabeled data has gained popularity in the transportation field. Its diverse applications include grouping public transport users based on the organization of their activities over multiple weeks, finding patterns of bus operation level based on stop frequency and stop duration in each hour, clustering passenger trip data based on mode choice, and finding patterns linking truck speed and segment traffic volume ( 19 – 22 ). Although K -means clustering is most commonly used for finding underlying patterns within large datasets, there is no established threshold for sample size, making K -means a viable tool for many applications.…”

Section: Methodsmentioning

confidence: 99%

Truck Parking Usage Patterns by Facility Amenity Availability

Mahmud

Akter²,

Hernandez

2020

Transportation Research Record

View full text Add to dashboard Cite

Truck parking is currently ranked by the American Transportation Research Institute (ATRI) as the fifth most critical issue for the trucking industry and, more importantly, as the second most important issue for truck drivers. Part of the problem can be attributed to inadequate supply of parking and federal hours of service (HOS) regulations. Recent truck driver stated-preference surveys reveal that amenities including restrooms, fuel, and showers are important considerations while seeking available parking. A link between parking usage patterns and facility amenity bundles can guide transportation agency investments in relation to the design and type of parking facilities with high potential to mitigate overcrowding issues, and can be used for predictive modeling in real-time parking availability algorithms and information systems. This paper used historical, anonymous truck global positioning system (GPS) data to determine the extent to which hourly parking usage patterns, that is, average parking duration, percentage of parked trucks, and parking usage ratio, vary by amenity availability. A K-means clustering model grouped parking facilities by time of day parking usage patterns, season, and geographic region. Each cluster, represented by parking usage patterns, was then tied to unique amenity bundles. Three usage pattern clusters were identified: overnight usage with long parking durations ( Cluster 1), off-peak usage with long parking durations, ( Cluster 2), and off-peak usage with short parking durations ( Cluster 3). In general, overnight and longer duration parking was associated with facilities that had fewer amenities, notably without showers, while peak and off-peak hours and shorter duration parking was associated with full-service facilities.

show abstract

“…An algorithm like k-means clustering [51] cannot be used satisfactorily here because the number of clusters or k must be predefined to proceed with the clustering. A few studies have applied the k-means algorithm to cluster public transit operations but suffered from the predefined k issue [52,53]. To remedy the shortcoming of the k-means, we adopted an approach known as X-means [54], which is a clustering technique that does not require a predefined number of clusters to proceed with the clustering.…”

Section: Route Clusteringmentioning

confidence: 99%

Visualizing public transit system operation with GTFS data: A case study of Calgary, Canada

Prommaharaj

Phithakkitnukoon

Demissie

et al. 2020

Heliyon

View full text Add to dashboard Cite

Public transportation agencies are one of the industries that generate a large volume of data on a high frequency and velocity basis. The General Transit Feed Specification (GTFS) is one of the datasets these agencies generate and share openly with the public. GTFS feeds contain data for scheduled transit service including stop and route locations, and schedules information. This paper aims to demonstrate the potential of GTFS data, specifically, the paper describes the development of a GTFS data visualization tool that displays spatial and temporal patterns of transit services from which qualitative information and insights can be gained. In this paper, GTFS data from Calgary Transit was used as a case study. Previous studies focused on the development of visualization tools that display transit movement, or static graphical representation of transit operation. However, there is still a need for a dynamic interactive visualization tool that can also measures and displays transit system operation geographically and statistically. This study builds on the previous investigations and further develops a new public transit system operation visualization tool (called PubtraVis) with six visualization modules that reflect on different transit system operational characteristics; mobility, speed, flow, density, headway, and analysis. The user can evaluate two modules side by side for comparative analysis. The analysis module provides an insightful statistical summary and similarity measure and clustering results based on the transit operation characteristics. PubtraVis was tested with real-world users through a user experience study from which it was found to be useful and easy to start using. PubtraVis can be a useful tool to demonstrate the dynamism of transit vehicles from the entire transit network at a glance, and can be used to facilitate communication between transit operators, city authorities, and the general public regarding the public transit planning and operation.

show abstract

Clustering of public transport operation using K-means

Cited by 11 publications

References 6 publications

Cluster Based Methodology for Scheduling a University Shuttle System

Cluster Based Methodology for Scheduling a University Shuttle System

Truck Parking Usage Patterns by Facility Amenity Availability

Visualizing public transit system operation with GTFS data: A case study of Calgary, Canada

Contact Info

Product

Resources

About