Analysis of SMART Traffic Management using BI/DM/KDD Techniques

This report examines the application of Business Intelligence (BI), Data Mining (DM), and Knowledge Discovery in Databases (KDD) in SMART traffic management systems. It reviews two articles: the first focuses on the STAR-CITY system, which uses spatio-temporal analysis to understand and predict traffic conditions, including journey time estimations and the integration of multiple data sources like weather and city events for traffic status prediction. The second article reviews the Dublin Dashboard, a web application that utilizes technologies like CakePHP, MySQL, PHP, JavaScript, and Leaflet to provide various modules and tools for traffic management in Dublin. The report discusses the system architecture, urban data used, and the key features of the Dublin Dashboard, including the use of interactive maps for visualizing traffic data. The analysis highlights how these systems leverage data to improve traffic flow, predict future conditions, and provide city managers with valuable insights. The report also references the use of OWL ontologies, semantic queries, and the integration of diverse data streams to enhance traffic management strategies.

Running head: SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
Name of the student:
Name of the university:
Author note:

1SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
Table of Contents
Article 1 review:..............................................................................................................................2
Spatio-Temporal Analysis of Traffic Conditions:.......................................................................2
Spatio-Temporal Analysis of Traffic Status:...............................................................................2
Spatio-Temporal Analysis of Traffic Context:............................................................................3
Star-city Traffic Status Prediction:..............................................................................................3
Article 2 review:..............................................................................................................................4
Dublin dashboard overview:........................................................................................................4
Technologies used in the application:..........................................................................................4
Key points of the article:..............................................................................................................4
System architecture and urban data:............................................................................................5
Dublin mapped:...........................................................................................................................5
References:......................................................................................................................................7

2SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
Article 1 review:
Spatio-Temporal Analysis of Traffic Conditions:
With the increase in the number of vehicles in the urban areas, transportation became o of
the major concerns in the cities. Congestion in the roads and immovability of the vehicles led to
commotion and it was getting increasingly difficult to control the traffic. The traffic managers
started implementing a system called the STAR-CITY, which stands for Semantic Traffic
Analytics and Reasoning for CITY, as a system, which integrates structured as well as
unstructured traffic data, velocity and volume of the historical data to understand the historical
and traditional issues associated with traffic management. The Spatio -Temporal Analysis of the
problematic traffic conditions stream calculates the journey times (estimation of travel time
between fixed
Points within Dublin Ireland). Fast aggregation techniques are used to process this data on a real
time basis (Nandhini & Shanthi, 2016). Historical analysis of the traffic status is done based on
the average, maximum and as well as the minimum time obtained. STAR-CITY helps in
discretizing travel time numeric values using roads directions as well as links in the status (Mika
et al., 2014). This is done using the SWRL5 rules. Temporal Selection of Date and time using
OWL EL++ ontologies as well as spatial selection is also done in the STAR- CITY model.
Spatio-Temporal Analysis of Traffic Status:
Retrieving the important contextual data from the widely spread city area is a quite
challenging task because the conventional technique of searching the data is has limited features.
It is only capable of catering to the needs of the city managers and handling only similar
incidents that has occurred in the past (Luo et al., 2017). It is also not much capable of handling

3SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
heterogeneous data recovered from different traffic conditions from different parts of the city.
The STAR-CITY system eliminates these issues and helps in proper searching of data.
Spatio-Temporal Analysis of Traffic Context:
The OWL 2 EL ontologies greatly affects the reasoning of STAR-CITY diagnosis, therefore an
OWL 2 distributed classification of EL journey times is used to obtain the scalable diagnosis.
The current technique of implementation is only limited to the scalability reasons of EL++
expressivity. By having a detailed knowledge of the current system context (spatial, temporal, as
well as traffic conditions), a semantic query is formulated to better identify the real information
needed by the city traffic managers within a particular traffic status or a city setting. The relevant
information like events as well as traffic conditions is then retrieved using a semantic search
(Kotoulas, 2014). It displays the results from an exploration interface to the traffic managers.
This helps the traffic managers to have a more clear insight into the traffic conditions of the city
as well as the impacts of the city events on the city traffic conditions.
Star-city Traffic Status Prediction:
Estimating future observations with historical information is used in the STAR-CITY
traffic management system by making use of the below explained formula:
- Integration of multiple sensors = heterogeneous data + exogenous data + raw data
streams such as information on weather, road works as well as city events or incidents,
and their accuracy as well as consistency in prediction.
- Result: Future status of the road segments in the future within the given boundary box
and their value of proportion that is reported maximum before 5 hours.

4SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
Article 2 review:
Dublin dashboard overview:
This article describes a web application known as Dublin Dashboard, which runs using a
webpage, and has 12 different modules (Damaiyanti et al., 2017). Different bespoke applications
as well as a collection of curated tools, which are relevant to the traffic systems in Dublin, are a
part of this application.
Technologies used in the application:
The application uses the framework of CakePHP11 for the MVC paradigm. A MySQL
database supports the same. CakePHP technology allows faster development by providing the
necessary scaffolding as well as implementation code for an efficient pattern of MVC. It also has
the MIT License copyright. PHP and JavaScript libraries are used to process the data for
providing the graphical user interface for the dashboard whereas high charts are used in
rendering the graphs of time series. Interactive maps make use of leaflets.
Key points of the article:
This paper describes the features as well as the functionality of Dublin Dashboard and the
available tools for usage in Dublin. This article speaks of the extensive audit of the available data
for the determination of quantity as well as the quality of Dublin datasets. This is the initial step
for any urban dashboard project. Later the technologies used in the dashboard were also
discussed which main namely java and php (Buisson et al., 2013). It has been explained how
applications can be effectively reused in order to ensure better efficiency and productivity form
the Dublin Dashboard. The open source mapping tools for the website were properly placed for
selection of the right technologies in order to develop the Dublin Dashboard. The framework

5SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
used in the development of the website also has provision of adding newer features to the
website in the future. It has been clearly explained in the article.
System architecture and urban data:
An architectural planning of the Model-View-Controller or MVC I is used to design the
dashboard which efficiently separates the data, the interface as well as the processing logic as
shown in figure 1 below (Adoni et al., 2013). The bespoke data elements from the Dashboard are
included by the web services in a periodic interval or they may also be downloaded and manually
stored in a file system. Users interact with the Dublin Dashboard via web browsers.
Figure 1: Dublin dashboard architecture as well as technologies (Source: As used by the author).
Dublin mapped:
Dublin Mapped is a feature that provides the complete set of maps showing results of the
two recent censuses of Ireland. Interactive maps are used to provide the data at a much smaller
level of area, which is equal to address values ranging from 80 – 120. The statistical data thus

6SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
obtained helps in mapping of the traffic in different parts of Dublin (Venter, 2017). Predictive
reasoning of the traffic can be achieved by correlating the congestion based on the parameters
such as rain, city events as well as road repairs etc.

7SMART TRAFFIC MANAGEMENT WITH BI/DM/KDD
References:
Adoni, W. Y. H., Nahhal, T., Aghezzaf, B., & Elbyed, A (2013). Conception of Intelligent
Transportation System using IBM Big Data Solution.
Buisson, C., Gillieron, P. Y., Heydecker, B. G., Lint, V. J., & Papamichail, I. (2013). The
NEARCTIS project: Network of excellence for cooperative traffic managment (No.
EPFL-ARTICLE-188475).
Damaiyanti, T. I., Imawan, A., Indikawati, F. I., Choi, Y. H., & Kwon, J. (2017). A similarity
query system for road traffic data based on a NoSQL document store. Journal of Systems
and Software, 127, 28-51.
Kotoulas, S. (2014, September). Semantic and Reasoning Systems for Cities and Citizens.
In Reasoning Web International Summer School (pp. 369-387). Springer, Cham.
Luo, X., Dong, L., Dou, Y., Zhang, N., Ren, J., Li, Y., ... & Yao, S. (2017). Analysis on spatial-
temporal features of taxis' emissions from big data informed travel patterns: a case of
Shanghai, China. Journal of Cleaner Production, 142, 926-935.
Mika, P., Bernstein, A., Welty, C., Knoblock, C., Vrandečić, D., Groth, P., ... & Goble, C. (Eds.).
(2014). The Semantic Web–ISWC 2014: 13th International Semantic Web Conference,
Riva del Garda, Italy, October 19-23, 2014. Proceedings (Vol. 8797). Springer.
Nandhini, K., & Shanthi, I. E. (2016). Analysis of Mining, Visual Analytics Tools and
Techniques in Space and Time. In Proceedings of the Second International Conference
on Computer and Communication Technologies (pp. 547-556). Springer, New Delhi.
Venter, K. (2017). Driver perception of non-motorised transport users: A risk in traffic?.