The shortest path algorithm performance comparison in graph and relational database on a transportation network

Miler, Mario; Medak, Damir; Odobašić, Dražen

doi:10.7307/ptt.v26i1.1268

Cited by 26 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Several researchers find that Neo4j doesn't support the route planning for transit transport [9,13]. Others used Neo4j only to store transit data [2,8,12].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Route planning on GTFS using Neo4j

Vágner¹

2021

AMI

View full text Add to dashboard Cite

GTFS (General Transit Feed Specification) is a standard of Google for public transportation schedules. The specification describes stops, routes, dates, trips, etc. of one or more public transportation company for a city or a country. Examining a GTFS feed it can be considered as a graph. In addition in the last decades new database management systems was born in order to support the big data era and/or help to write program codes. Their collective name is the NoSQL databases, which covers many types of database systems. One type of them is the graph databases, from which the Neo4j is the most widespread. In this paper I try to find the answer for the question how the Neo4j can support the usage of the GTFS. The most obvious usage of the GTFS is the route planning for which the Neo4j offers some algorithms. I built some storage structures on which the tools provided by Neo4j can be effectively used to plan routes on GTFS data.

show abstract

“…Several researchers find that Neo4j doesn't support the route planning for transit transport [9,13]. Others used Neo4j only to store transit data [2,8,12].…”

Section: Discussionmentioning

confidence: 99%

“…Miler [13] stated that "graph database management systems are not routing engines and are not suitable for full graph traversal, which is used in the shortest path calculations". They also said that "if the memory is not an issue, then graph database is the right choice for the shortest path calculation.…”

Section: Related Workmentioning

confidence: 99%

Route planning on GTFS using Neo4j

Vágner¹

2021

AMI

View full text Add to dashboard Cite

show abstract

“…Consequently, it would be interesting to find other problems that could be solved by this high performance algorithm, e.g., A * and Dijkstra. Using a proper algorithm like the one in pgRouting enables even a RDBMS like PostgreSQL to show good performance for shortest path queries so that Neo4j was only 20-40% faster depending on cache configuration and only because Neo4j had enough memory for caching [56]. A comparison between traversals using Neo4j's own traversal framework and manual traversals showed only minor differences in performance with no clear advantage for any approach.…”

Section: Algorithms Usedmentioning

confidence: 92%

Querying a graph database – language selection and performance considerations

Holzschuher

Peinl

2016

Journal of Computer and System Sciences

View full text Add to dashboard Cite

“…In this study, stations are regarded as nodes, and sections are denoted by links, similarly described in [15]. Notations of the urban rail network are as follows: …”

Section: Notationmentioning

confidence: 99%

A Key Station Identification Method for Urban Rail Transit: A Case Study of Beijing Subway

Sun

2017

PROMET

View full text Add to dashboard Cite

INTRODUCTIONWith the rapid development of urban rail transit network operation technique, the increase of passenger volume leads to congestion in stations [1], causing even safety risk in the network [2]. Besides, the topology of urban rail transit network can also aggravate or relief the congestion situation of subway networks. Therefore, an effective method identified the key station is an imperative to manage the overcrowded urban rail transit.For the factors of the station performance, many of the current approaches focus on the relationship between passenger flow and facility capacity. Xu et al.[3] combined queuing theory with facility utilization evaluation to conduct the station capacity evaluation, according to the gathering and scattering process of a station. Kepaptsoglou and Karlaftis [4] proposed a model to evaluate passenger evolution on subway station platforms, which usually gathered most passengers. Wang et al.[5] also used the passenger evacuation capacity to evaluate the station based on the facility utilization. If the service level of the facility is too low, there are big problems in the station.Besides, passengers' subjective feelings and other factors are also used for evaluation. Matulin et al.[6] proposed a two-level evaluation method to analyze public transport performances from micro and macro levels, using performance indicators and quality of service indicators. Kesten and Öğüt [7] developed a measurement framework of passenger-oriented performance to evaluate rail transportation, using satisfaction and importance ratings obtained from passenger surveys.Meanwhile, there are many approaches employed in the station evaluation. Some typical methods such as Data Envelopment Analysis (DEA) have played their great role in the station assessment. Sancha et al. [8] used the method to estimate the technical, social and environmental efficiency of transit transfer stations. Pjevčević et al. [9] used DEA method to measure and analyze the efficiencies of ports on the Danube River, which can be borrowed for station evaluation. In addition, simulation is also another method, where relevant simulation or operation data are provided. Zhang et al. [10] proposed a cellular automata-based model to simulate passengers' alighting and boarding behaviours in metro stations, aiming to analyze and evaluate the operation performance of stations. However, most works above do not consider the influence of other stations in the subway system, which results in local rather than global conclusions.Fortunately, in recent years, a complex network theory has been introduced to the station evaluation from a global viewpoint. Indexes such as degree centrality, betweenness centrality, and closeness centrality have an outstanding ability to identify key nodes from the network [11][12][13][14]. Sheikhahmadi et al. [12] proposed an improved high degree centrality index to identify the Traffic Management Original Scientific Paper

show abstract

The shortest path algorithm performance comparison in graph and relational database on a transportation network

Abstract: In the field of geoinformation and transportation sci-

Cited by 26 publications

References 14 publications

Route planning on GTFS using Neo4j

Route planning on GTFS using Neo4j

Querying a graph database – language selection and performance considerations

A Key Station Identification Method for Urban Rail Transit: A Case Study of Beijing Subway

Contact Info

Product

Resources

About