Mohandu 2021
Mohandu 2021
Mohandu 2021
a r t i c l e i n f o a b s t r a c t
Article history: Big Data is an evolving exemplar moreover, particularly within in the transportation system, has now
Received 7 March 2021 become a powerful enticement of global interest. On the basis, Big Data is often seen intact undertaking
Accepted 17 March 2021 whereas transit system fore efficiently handle utterly the information needed by aforementioned part to
Available online xxxx
provide secure, unstained including reliable modes of transit, and to create custom the recipient’s trans-
port experience. By having access to centralized services, a smart city enhances the effectiveness of its
Keywords: inhabitants. Intelligent Transportation Systems (ITS) play an essential role in the transformation of a cos-
Big data analytics
mopolitan environment into a digital city. With the last two decennary, numerous ITS technologies have
Smart transport
Intelligent transport system
been deployed, such as city-wide traffic regulation control, smart parking, general transport information
assistance (transit vehicle, rail, compact, air, etc.), shipping, legitimate traffic, measuring of highway
speed restrictions, etc. Most subsequent Data Analytics insight is needed for transport and movability
industry, ITS implementations, threshold instances, areas and use cases, including the routing, planning,
ability to track of infrastructure, platform architecture, and more, are examined in this article. Finally, the
paper tackles some open problems in ITS with the use of big data analytics.
Ó 2021 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the 12th National Confer-
ence on Recent Advancements in Biomedical Engineering.
1. Introduction data tools and intelligent control techniques, are integrated even
in transportation networks upon the ITS [3].
In both academia and industry, Data analytics is now a debating Objective of ITS is to provide good infrastructure to road users
point. It contains massive data out-of various sources. Many of the and riders in transport systems. Information collected diverse
Big Data mechanisms include text mining, data science, robotics, sources in the ITS, for instance cash plus card, GPS, actuators, mul-
socioeconomic networking, knowledge discovery and much more. timedia locater, internet community, etc. using detailed efficient
[1]. For economies and characteristics of life around the world, suc- content analysis with apparently undisciplined records, the infor-
cessful expansion of cities and automobiles is always important. mation created in ITS expanding Trillion byte to Petabyte with
Inefficiencies are costing money, rising emissions and taking time expansion of ITS. Conventional database frameworks are inade-
out of the lives of people. The problem is that transport infrastruc- quate, with this volume of data, and do not fulfil the requirements
ture supplies are rising shorter unlike trend. Automobiles has been for data analytics. This is because the exponential growth of the
designed quicker than tracks. Urban areas are growing smarter like volume and sophistication of data is not foreseen. Big Data analyt-
a roadway, but it seems the automobile industry is shifting to data ics offers a modern technological approach for ITS. ITS may benefit
analytics to search alternative ways to existing capital, reduce con- from the analytics of Big Data. Large data systems including HDFS
gestion and enhance the experience of travel. Since the start of the and Spark’s, which are being widely used in academic and com-
1970s, intelligent transportation systems (ITS) have been built. mercial, are capable of processing large quantities of data. Top
That’s the long term future including its transportation network. social media sites like Facebook, Twitter, Wei bo, and We Chat deli-
Advanced technologies, including automatic innovations, wireless ver citizens with pervasive opportunities through exchange
thoughts, emotions and knowledge officially or in learning associ-
ations, producing enormous amounts respecting concurrent gen-
eral waving [5].
⇑ Corresponding author. Social network collects the mortal portability data through
E-mail addresses: [email protected] (A. Mohandu), [email protected] (M. smartphones, and even evoke vast number with social signals.
Kubendiran).
https://doi.org/10.1016/j.matpr.2021.03.479
2214-7853/Ó 2021 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the 12th National Conference on Recent Advancements in Biomedical Engineering.
Please cite this article as: A. Mohandu and M. Kubendiran, Survey on Big Data Techniques in Intelligent Transportation System (ITS), Materials Today: Pro-
ceedings, https://doi.org/10.1016/j.matpr.2021.03.479
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
Compact and flexible equipment (e.g. smart phones) can access 2. Background
information [6]. During this previous assertion, we will propose
that the new field of social transportation should concentrate pri- Smart transport’s comprehensive development draws analyzer
marily on five areas: 1) big data and social signal traffic or trans- often in context of an insightful architecture of Intelligent Trans-
portation analytics using pattern discovery, data science and portation System. A variety of benefits can be taken forward by
speech recognition tactics; 2) Publicizing frameworks upon social the regular smart transport system. Investigator move currently
media, Personalization, the Internet of Things (IoT) or perhaps engaged to define the overall result with intelligent transit in order
(IoT) based transportation mechanisms. 3) New location-based ser- to attain meaningful computing [13]. Similarly, in the current liter-
vices (LBS), like automation of transit expertise, in particular ature, a range of suggestions have been presented that seek rigor-
decision-based services (DBS) or task-based services (TBS) compre- ous experimentation and simulations to overcome the issues based
hensive information needed through timely manner for transit on the test bed. Whenever the concerns regarding commuter traffic
actions or objectives, but instead communication or intelligence being examined, therefore concentrate on traffic density, remains
based services (IBS) or knowledge-based services (KBS) which sug- factors such as traffic speed, road etc. are also considered to affect
gest operators or activist groups that may obtain information. 4) traffic. Freight performs an identical decisive aspect in every soci-
Web-based mobility regulation and strategic agent technology ety earnings including every resident growth. This article explains
[8], other than intelligent machines, information automation or the technique of reducing RFID congestion problems instead of the
app advisers besides monitoring system, preventative mainte- conventional adhoc gateway method. It is also clean and cost-
nance, health and energy governance, should seek to develop dif- effective. The aim of aforementioned procedure intend to manage
ferent IoT devices which really accumulate community road and track roadblock in the event of a necessity.
connect individuals with congestion and automobiles through Smart transportation based on IoT adoption certain data analyt-
legitimate time; and 5) actual solutions or even reviews towards ics frameworks do important whereas information refine, such as
further growth as well as analysis. Incorporating commute evalua- MapReduce, Cassandra, NoSQL, etc. For information analysis,
tions, each linear motion model and perhaps cognitive science, Hadoop uses the Map Reduce mechanism [14]. MapReduce the
researchers have modelled various aspects of transport. It is impor- job in2 distinct phases: (1) mapping. (2) reduce. The mapping task
tant for advanced analytics structures that provide a distributed convert integrate information gathered in the mapping in a modi-
design that mould simple toward build along with address when- fied ways used to reduce the process. MapReduce produces several
ever a problem occurs. data sets that are responsible for a comprehensive selection of rou-
Advanced Analytics be comic method about analyzing vast tine activity. The information is separated by MapReduce within
quantities of data containing aforetime information categories triv- autonomous are handled simultaneously. The framework pattern
ial unknown trends, hidden equate, retail fashion, consumer trends classifies the outputs (map) and transfer them toward following
or other business intelligence evidence; [23]. Classification of ana- assignment (reduce). These tasks are performed simultaneously
lytics, namely, descriptive analytics, predictive analytics, and per- by MapReduce, i.e. integrity, error leniency, and load distribution.
spective analytics. Descriptive analytics that describe ‘‘what has Society now turn to innovation mechanics and intelligent transport
happened or what is going on” help to discovery modern event methods are mere informative. Events move related also expressed
and threat for business. Predictive analytics that specify ‘‘What’s along one another via the internet and this vision of the internet is
going to happen and why”, facilitated applying multiple techniques assumed to be Ubiquitous IoT [16]. In this literature, found to man-
[24]. Prescriptive analytics, that specify ‘‘what should I do and why age a massive volume of data as well as provides IoT applications
should I do it,” refers to the systematic of Visualization, compe- to a diverse customer base. However, massive Data bear different
tence as well as analyzing through examine different preference impenetrability, IoT is beginning a valuable examine analysis. In
and implement decision makers with proposal. To predict addition, the refine of massive data and the direction of the enor-
exactly future circumstances and states, like text/web/data mining mous online-offline input against the IoT-based domain are used
[25]. by different architectures.
The volume and speed by transport and movability knowledge
endure produced today has far exceeded the scales at which at the
beginning of this century they used to be collected, processed and 3. Smart city and smart transportation
analyzed. The combination of modern digitalization ideologies
including Internet of Things (IoT), the rise of digital Communities’, 3.1. Smart city
the sharp decline in data transfer expenses of sensor industry, the
economical also popular use of sensors or personal computers, A huge complex system has been created by the brisk expansion
each issues raised in digital networks, increased drastically like different metropolis aspects involve, with digital government,
enhanced the intelligence of humans facing gain further compre- IT proposal, machinery, framework, and habitat. Such a compound
hensive understanding of travel. A transport network persist sev- structure presents a range of threat and liability, with examples
eral platform that enables individuals with different aspects vary from carbon emissions and blockage, to rising unemployment
include automobile, bus, train, walking, bicycle, etc. to travel from rates and negative social impacts. The use of information and com-
an origin to a destination. Despite the accessibility and connectiv- munication technology (ICT) to make cities ‘‘smart” is one solution
ity advantages of transport networks, as the usually-increasing rise for managing urban problems and improving the livability, oper-
in community but rather volume of traffic, cities have always been ability and sustainability of cities. There are three layers in the
dealing with mobility concerns (e.g., traffic congestion and air pol- specific context of intelligent community: 1) data acquisition and
lution). Brilliant transportation, as a savvy and huge segment of the manipulation, 2) systems analyst, 3) resource provision. First, data
shrewd civic, carry emotional auspicious objective to ease concern must be collected with the help of smart devices located through-
and enhance urban communities live ability, usefulness, and main- out the city, such as traffic sensors, utility usage sensors, weather
tainability by creating astute and novel vehicle models [12]. A stations, mobile phones, and social media networks, when building
definitive viability of keen transportation models is profoundly a smart city. In order to prepare temporal information, heteroge-
dependent upon the nature of sent information and qualities of neous data from different sources must be organized using data
the insightful methods into which the information are taken care normalization and data management techniques as the main
of. inputs for the analytics phase are shown in Fig. 1.
2
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
3
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
sification of human activities [18]. The surveys relevant to our rapidity, since this acceleration is analyzed by the OBU [20]. The
paper are further developed in Table 1. A summary of each relevant spacing is inferred from the vehicle’s spot (vLoc) and the signal
survey paper’s main focus is given. It is clear that, although there point’s location down the lane. The GPS encoded as in vehicle cor-
are a number of data fusion surveys, these are mainly focused on respond the sensor values, whereas position of signal point is
areas or classifications of specific applications (e.g. ITS embedded determined using corresponding Lane Id (lanId), about every pay-
sensors in mobile devices). load heading against RSUs, the mobility datasets are kept updated.
4.3. Evaluation of data analytics techniques in C-ITS 4.4.2. Destination list dataset
Destination List dataset (dList) Specifies a data source repre-
The Cooperative Intelligent Transport (C-ITS) System endure an senting all the different destination locations (desLoc) that the
essential facilitator of inevitable state traffic regulation. An innova- vehicles are proceeding. The riders sustains device location in the
tive C-ITS study implemented automobile, street side entities and vehicle OBU either at beginning, so every new service from RSUs
control zones for traffic they produce a lot of traffic, that incorpo- is reviewed mostly with RSUs travel desired place and including
rates contextual movability and infrastructure. [19]. In C-ITS, data mobility data mostly from destination list (dListData).
analytics tools can be used to assess the accuracy of services by
means of congestion control and competitive transmission. Certain 4.4.3. Traffic flow dataset
C-ITS services that mostly depend on intelligent choices informa- Traffic flow dataset (tFlowdata) organizes information in a tra-
tion gathered, a few data analytics strategies often optimize accu- vel time database from different LDSs (tFlowdata), lane id (lanId)
rate simulation in Big Data computing. Several more C-ITS but the differing volume of vehicles are included (vCount). A loop
technologies obtain and examine but provide user-based data. detector for stationary vehicles capture levels of automation resid-
Two such applications are discussed here, namely the smart park- ing at the frequency (LDS). After each period of time, the traffic
ing system and monitoring of road conditions, study offers an flow data (tFlowdata) is retained up to date.
intelligent, cloud-based parking via the Internet of Things (IoT).
Thus every car has quite an RFID tag and RFID reader is accessible 4.4.4. Pre-decision datasets
there in car at both entry and exit points. The system is imple- Pre-decision Datasets (pDecData) cloud aspect of Time Calcula-
mented collects data on approaching traffic employing RFID read- tor (TC) quantify time to reach (TTR) of each automobile using cor-
ers. Users allows to book a parking space with a mobile app until a responding mData refer the corresponding sensor abandoned. The
database controller intends to assign parking space. obtained values track-wise pre-decision (pDecData) time-series
embedded in various pre-decision data repositories (PDD) follows
4.4. Cloud storage TTR is determined.
Cloud based data is quite well categorized in online storage that 4.4.5. The Navigator (NG)
attains the cloud. Data Manager (DM) is responsible for classifying Cloud module takes dListData but the online service to deter-
the different datasets referred on the basis of their sources. The mine all optimal routes with each dListData destination. A new
Time Calculator (TC) and Navigator (NG) of cloud elements utilize dataset, a sequence about all the adjacent nodes to associated
data sources labeled by DM [32], to build innovative data points in places, is deposited there in shortest path (sPathdata) for every
cloud services. The data sets operated in cloud storage were indeed registry, comprised of a desired destination even a specify of all
illustrated as follows. the nearest neighbors along their intervals, contribute to subse-
quent given destination. To determine all quickest pathways to
4.4.1. Mobility datasets the route, NG benifit Dijkstra’s single-destination shortest path
Mobility datasets (mdata) describes RSU data based on their algorithm.
Lane Idd in different datasets (lanId). These datasets of distinct
routes are processed in the Mobility Data Repository (MD). The 5. Literature survey
mdata append vehicle id (vId), vehicle velocity (v), vehicle air resis-
tance (a), vehicle location (vLoc) and vehicle distance (d) from lane The premise of big data has been interpreted throughout the
joining communication phase. On-board sensors capture the framework of transit and mobility, a research study of smart
Table 1
Comprehensive review of work associated with data analytics in Intelligent Transportation System.
Legend: CC = Cloud Computing, IoT = Internet of Things, BDA = Big Data Analytics, ITS = Intelligent Transport System.
4
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
Table 2
A summary of existing primary work focus on Application.
transportation systems would be provided, emphasizing the diver- The ensuing inventory outlines in designing 3 majority important
sity of existing literature in recent decades, putting forward a ques- marker that distinguish large data set from typical data.
tion of route map for deploying large traffic data on the cloud
(Table 2). 5.3. Data-fusion architectures
5.1. Big data in transportation and mobility Attempt to acquire a deep assessment of flexibility behavior and
traveler trends which cannot be extracted from a given criterion,
Big Data is coming to transportation industry to solve conven- data fusion algorithms are employed to individual but supplemen-
tional digitize issues to provide modern scope, application mainte- tary data sources. The main objective of smart transportation data
nance through unique data origin and applied science. The fusion algorithms to establish an advanced automation strategy
legitimate implementation of traffic influencing data streams is that connects expertise from disparate pieces of knowledge. The
indeed ideal example; such a specific problem is not revolutionary, types of input/output, which are subdivided into different data/fea-
but its volume, velocity and range seem to be attributes which ture/decision combinations versus data/feature/decision output. In
compromise the information processing. Contextualized descrip- fusing techniques, the protocol layers of the data installed, where
tion towards Big Data mixes collectively assortment of huge mea- raw data, features, and decisions are three levels of abstraction that
sure of immense speed, heterogeneous, developing area are transformed by feeding into fusion algorithms to the more pre-
information and utilization of cutting edge methods and exem- cise levels. In the transportation system [32], Big Data comes from
plary to repository, recover, oversee, measures break down the many sources, such as road and inside vehicle traffic monitoring
caught data [26]. systems. In addition, during normal, uncertain or crash conditions,
Big data means large quantities, instant rather than realistically detector set up in automobile play a significant role in analyzing
provides context infrastructure demanding multiple categories of information. In addition to the network architecture, technologies
optimizing complexities, the inclusion with expertise and process such as Global Positioning System, mobile, WPAN, Radio Naviga-
simplification. From the point of view of transport, apparent as a tion, Automatic Vehicle Identification performs significant job.
listing of applied science enable all the data needed via progress
different path of providing impregnable, further effectual transport 5.4. Collection of transportation and mobility data
to be managed effectively (store, process and access), representing
consumer to illustrate and customize transport familiarity. Three Consequently big data gathering are being comprehended as
transportation features and scales are major aspect quote in refer- from the assignment of accumulating data gathered by sensing
ence to Big Data characteristics 3 Vs volume, velocity and variety devices that have been triggered via various transportation and
model defined by Gartner. There are other aspects, such as veracity, mobility platforms by residents and aircrafts. Information gather-
visualization and visibility, referred to by other authors. ing comprises a number of assignments to obtain data from fre-
quencies. Implementation able to contribute of the actual data
5.2. Features and scales of transportation Big data structure even these electronic data, specimen which can be orga-
nized via a computer within electronic signals.
Additionally, we will concentrate on the previous variables, as Recently, Data Analytics in transit and movability, global tech-
they are important in the manner in which information is col- nological trend recognized as elite rate of model for which pattern
lected, analyzed, organize, repository, diffused and treated [29]. assumption, learning and optimization system are used.
5
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
Since the present stage, equivalent instruction of MapReduce different broadcasts facility besides the respective exchange of
and HDFS will be addressed. In addition to HDFS, SQL’s supposed security non-security information.
HBASE, HIVE need into store historical data for database adminis-
tration (offline or in-memory). In the Hadoop cluster, MapReduce 5.8.2. Data diffusion
and HDFS overall function. Consequently, we also need a data Additional challenging task connection with information
stream in real time as well as offline data analysis. In addition, insights is successful data diffusion. Automobiles accept recurrent
we need a 3rd party software to combine Hadoop computing pro- CAMs for making choices related to security implementations, as
tocol for legitimate computation in order to deliver implementa- discussed earlier. As assumptions on reliability are taken on basis
tion in actual environments. Spark is used for legitimate of community movability information, these messages must be
implementation, mostly with the Hadoop platform. For extensive received with high reliability by vehicles. The default wireless
data processing, Apache Spark is a specific engine. It provides Big technology for the dissemination of these messages within the
Data with faster access and allows recyclability across streaming local community region is IEEE 802.11p in this context. Along with
applications. Usually used for big data workloads, interface of large matrices, from the instance alert messages, might be used a
advanced parallel computing. Fast performance, combines buffer- multi-hop approach respectively vehicles and RSUs may be
ing and enriched effectiveness in cognition. Apache Spark is informative.
assisted by Amazon on Hadoop YARN patterns are being checked
and performed quickly. 5.8.3. Data calculation
ITS data calculation can be executed with a distributed or cen-
5.7. Big data approaches used in ITS application tralized way. Instead every driver preserve to evaluate automobiles
have been associated with a virtualized environment but rather
Now it is possible to access and share detailed transport data on communicate periodic Recorders, results to give itself a safety deci-
unprecedented scales, which reveals new paradigms of transporta- sions including utilizing brake system, oncoming traffic includes
tion and smart mobility advantages. A multitude of transport and several interventions.
mobility applications are supported by the Big Data framework,
which includes some including all of the key stakeholders in the 5.8.4. Metrics
study entry:[34] Residents (walker), the civil service (executives) Using effective metrics is an additional question with using C-
and capitalism. The encompassing of massive data approach proce- ITS analytics. In evaluating a reliability on safety applications in a
dure in real applications derived from data acquisition, observation C-ITS, primary consequence is identifying the level of consensus
and manipulation does not vary significantly between systems and of automobiles and its nature by enhancing effective safety mea-
services associated with automotive sector. Conversely, if there are sure on automobile integrity may be assured metric.
changes in the findings as well as scientific strategies incorporated
for those technologies, each stakeholder’s commitment will cover 5.8.5. Data analysis
the full massive data phase, from research process to its assertion Concise data analysis is another significant aspect of data ana-
with intellect records (that had importantly provoked a new set of lytics. Even more information could be received by RSUs and
issues devised with distributed computing. TCC, the volume of automobiles and expected amount of traffic
on the road and essential security algorithm stronger key lengths
5.8. Data analysis issues and implementations implies secure automobile methodologies, suitable inferences can
be made on such data. A k - mean clustering, correlated transition
Therefore, in this category experts evaluate several essential will result in a drop as in data transmission unless the volume of
questions pertaining to C-ITS data science. Such difficulties and the automobile is maximum than every interval. Within city, fuzzy
their potential solutions are discussed below [35] (Table 3). inference principles can been required to develop effectively trav-
elling determinations of essential contemporaries.
5.8.1. Data collection
Each C-ITS station collects data based on information on traffic 6. Open challenges
and mobility. Adequate originate of continuum 5.9 GHz as the
exchange of C-ITS relevant information has been assigned by both The substantial interaction evinced within that existing litera-
the IEEE and ETSI norms. The DSRC is again split in to communica- ture conducted around in the subsequent section could be a clear
tion link but a dedicated short-range communication continuum sign of the innovations recently introduced with data analytics
7
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
Table 3
Study of ITS data analytics technologies.
implemented to urban mobility areas. In addition, indeed a variety day, particularly ITS will have a huge range of information from
case studies hardly unresolved but never identified with moment, various actuators. Conventional database facilities and server
and yet major innovations persist also emerging a consequence of equipment were not capable of dealing via an excessive lot of
data acquire and impoverishment all over heterogeneous transport issues along with complicated records infrastructure and database
regions, adaptation to the latest Big Data features and actual data tools. [35].
analytics. Creativity as well as performance were further regulated Implementing the most appropriate memory structure is now a
while being consistent in providing actual data refine and applica- crucial issue for cloud services like Google, Microsoft are continu-
tion development. ing through enhance everyone assistance in a built-in efficiency of
data analytics Cloud assets, Solid State Drives storage are growing
6.1. Security and privacy as major aspects for huge information storage.
Data security is a long-standing issue that is widely studied in 6.5. Data processing
the fields of mobility and transport. In this respect, the survey pro-
vides guidance for future research in which the use of sensor data Predictability is essential for ITS Real - world applications, along
to improve battery power and time is proposed for smart e-bike with traffic data classification, traffic status identification, actual
control system. The statistical analysis was done and merged to traffic control, vehicle routing guidance and the scheduling of
ensure effective usability purposes along with approval organize actual public transport. Traffic data using multiple platforms via
aggregate advice needs in this area. a lot of methods can be evaluated to historical data and Analyzed,
with a minute period of moment. Data acquisition method is cap-
6.2. Data collection able of processing more complex and increasingly extensive data
[36]. Maintaining the reliability of the process with these strong
The data collected during mass transit could be inconsistent, and powerful statistics is a critical concern. A range of different
inadequate, or incompetent, despite the enormous momentum of massive data architectures which deal with actual datasets, like
vehicles and pedestrians, in particular locations perhaps at specific Apache Storm, Apache Flink, Apache Samza, Apache Spark Stream-
intervals. Instance, hardly complete automobiles will be integrated ing and Kafka Streams gained prominence.
with in key quality that implement actual position data and road
traffic statistics against traffic detector could be ignoring. Another 6.6. Data opening
conceivable solution to evaluate the contest will be to improve
existing quantitative strategies to improve data collection capabil- Information will be modified and shared with the public in high
ities with the adoption of IoT, promote specific detector tools quality to allow consumers of mass transit content providers to
through manually that can significantly improve information gath- consider reusable relevant data. The integrity of the information
ering and efficiency. pertains with its reliability, truthfulness, data analytics may delude
decision-making or even generate information without better
6.3. Data privacy quality performance with negative impacts. Setting up data of good
quality however may necessitate moment as well as payments. In
The most demanding and fretting issue in the Big Data era is respect, swap among establishing information sensitive to a
privacy. During data transmission, storage and individual informa- change expenses and creating qualitative information applicable
tion is sometimes revealed. Collected information from transport on massive cost, making this a difficult task to open up good qual-
systems, governments should review current data confidentiality ity data [37]. Technical programs provide the acceptance of artifi-
policy which comprises personal information are being reported cial intelligence for test execution by automatic data acquisition
with context of manuscript or use of records the general rules for and/or use. In addition, to ensure vibrant and accurate data on
dataset and sharing of resources and several aspects to avoid addi- transport entities might have implemented a trade activities.
tional loss of relevant sensitive data.
6.7. Big data analytics
6.4. Data storage
Infrastructure methodology is often applied in analytics unit.
At present, the volume of data stepped against Terabyte to Peta- The ITS input dataset is transferred into portions in the map-
byte, advancement in computing capacity is still initiating. Each reduce framework, therefore the mapper clusters are analyzed.
8
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
Table 4 has been explored. A lot of ITS Data Analytics platforms are being
Data Analytics Issues in ITS. addressed, along with route optimization, traffic congestion fore-
Reference Data Analytics Issues in ITS casting, fatal crash analysis, massive transit organizing, private
Ghanbari [4] 1- Vertical and Horizontal Scaling route planning and transport infrastructure management. Smart
2- Data Normalization mobility framework is developed for actual storage employing
3- Data Representation Advanced Analytics tools. We are indeed enhancing the tasks by
El Faouzi [18] 1- Integration scheme of trustworthiness and building a strong automobile data assets in place to ensure real-
dependability
2- Evaluation of Integration scheme with key
time decisions on real-world live traffic.
metrics
2–1 Adaptive nature of MOPs.
2–2 Relying on MOPs cluster based context.
Declaration of Competing Interest
2–3 Ground truth
3- File systems as well as promotional system The authors declare that they have no known competing finan-
software. cial interests or personal relationships that could have appeared
4-Design for worst-case scenarios.
to influence the work reported in this paper.
Nandury, 1- Actual traffic information involvement with
Satyanarayana [19] knowledge review
2- Strategies to handle Big data
3-Supervised Learning Model
References
4-Elementary Tools Assessment Perceptron and
Smarter Modem [1] J. Zhang, F.-Y. Wang, K. Wang, W.-H. Lin, X. Xu, C. Chen, Data-Driven Intelligent
Transportation Systems: A Survey, IEEE Transactions on Intelligent
5-Data Fusion Engine
Transportation Systems. 12 (4) (2011) 1624–1639, https://doi.org/10.1109/
6-Decision Logic
TITS.2011.2158001.
7-Smart Decision & Control Enabler
[2] O. Tene, J. Polonetsky, Big Data for All: Privacy and User Control in the Age of
Torre-Bastida [32] 1- Innovative revolutionary kinds of transit. Analytics, Northwestern Journal of Technology and Intellectual Property. 11
2- Prospect of transit in urban communities (2012).
3- Infrastructure upon motility [3] M. Shtern, R. Mian, M. Litoiu, S. Zareian, H. Abdelgawad, A. Tizghadam, in:
4- Modern Intelligent Transport system Towards a multi-cluster analytical engine for transportation data, IEEE, 2014,
5- Economic and Regulatory trends. pp. 249–257.
[4] R., Ghanbari, H., Zareian, S., Shtern, M., Litoiu, M.: A data platform for the
highway traffic data. In: 2014 IEEE 8th International Symposium on the
Maintenance and Evolution of Service-Oriented and Cloud-Based Systems
Mostly on assigned records, thus every mapper module performs (MESOCA), pp. 47–52. IEEE (2014)
independently, intermediate outcomes remain generated. Includ- [5] Bello Orgaz, Gema & Jung, Jason & Camacho, David. (2015). Social Big Data:
Recent achievements and new challenges. Information Fusion. 28. 10.1016/j.
ing modifiers which operate the clustering, the result obtained inffus.2015.08.005.
by reducers were integrated into final outcome which carried onto [6] C. Wang, X. Li, X. Zhou, A. Wang, N. Nedjah, Soft computing in big data
simulation component, while the result obtained by the mappers intelligent transportation systems, Appl. Soft Comput. 38 (2015), https://doi.
org/10.1016/j.asoc.2015.06.006.
are essential. [38]. [7] W. Yuan, P. Deng, T. Taleb, J. Wan, C. Bi, An Unlicensed Taxi Identification
Model Based on Big Data Analysis, IEEE Trans. Intell. Transp. Syst. 17 (2015) 1–
11, https://doi.org/10.1109/TITS.2015.2498180.
7. Discussion and future research directions [8] L. Rodríguez, C. Rodriguez-Enriquez, J. Sánchez-Cervantes, J. Cervantes, J.
García-Alcaraz, G. Alor-Hernández, A general perspective of Big Data:
applications, tools, challenges and trends, The Journal of Supercomputing. 72
The constraints of existing research are noted out in this sec- (2015), https://doi.org/10.1007/s11227-015-1501-1.
tion. As a consequence, few essential issues could arise to be [9] Engelbrecht, Jarret & Booysen, M.J. (Thinus) & van Rooyen, G-J & Bruwer, F.J..
addressed in future research support full use of big data trans- (2015). A Survey of Smartphone-based Sensing in Vehicles for ITS Applications.
IET Intelligent Transport Systems. 9. 10.1049/iet-its.2014.0248.
portation [22]. [10] X. Zheng, W. Chen, P. Wang, D. Shen, S. Chen, X. Wang, Q. Zhang, L. Yang, Big
Information gathered by ITS mechanisms is intended to opti- Data for Social Transportation, IEEE Trans. Intell. Transp. Syst. (2015), https://
mize opportunities with more extensive traffic congestion, further doi.org/10.1109/TITS.2015.2480157.
[11] Q. Shi, M. Abdel-Aty, Big Data applications in real-time traffic operation and
exploration is meaningful. Most of the recent findings depend with safety monitoring and improvement on urban expressways, Transportation
1 or 2 of a substantial dataset for transport, while certain numer- Research Part C: Emerging Technologies. 58 (2015), https://doi.org/10.1016/j.
ous factors of valuable insights. [39]. trc.2015.02.022.
[12] Vlahogianni, Eleni. (2015). Computational Intelligence and Optimization for
Data on transportation is being obtained through various chan- Transportation Big Data: Challenges and Opportunities. 10.1007/978-3-319-
nels. Even so, there has been highly heterogeneous data. Data 18320-6_7.
aggregation would therefore throughout the longer term serve a [13] Y. Xia, J. Chen, X. Lu, C. Wang, C. Xu, Big Traffic Data Processing Framework for
Intelligent Monitoring and Recording Systems, Neurocomputing. 181 (2015),
powerful transport systems of data analytic effort.
https://doi.org/10.1016/j.neucom.2015.07.140.
Simulators of urban transit systems possibly vastly upgrade [14] Rathore, Muhammad Mazhar & Ahmad, Awais & Paul, Anand & Thikshaja,
upon basis of actual data via a smart transport collected device Uthra. (2016). Exploiting real-time big data to empower smart transportation
[41]. Aforesaid program for simulation a significant role in ramping using big graphs. 135-139. 10.1109/TENCONSpring.2016.7519392.
[15] X. Wang, Z. Li, Traffic and transportation smart with cloud computing on big
up ITS advancement whereas limiting new platform expenditures data. 13 (2016) 1–16.
being tested. [16] I.A.T. Hashem, V. Chang, N.B. Anuar, K. Adewole, I. Yaqoob, A. Gani, E. Ahmed,
Some of the Big data analytics challenges are expressed in H. Chiroma, The Role of Big Data in Smart City, Int. J. Inf. Manage. 36 (5) (2016)
748–758, https://doi.org/10.1016/j.ijinfomgt.2016.05.002.
Table 4 [17] Hua, Xingcheng & Wang, Jierui & Lei, Li & Zhou, Bin & Zhang, Xiaolin & Liu,
Peng. (2016). H-TDMS: A System for Traffic Big Data Management. 626. 85-96.
10.1007/978-981-10-2209-8_8.
8. Conclusion [18] El Faouzi, Nour-Eddin & Klein, Lawrence. (2016). Data Fusion for ITS:
Techniques and Research Needs. Transportation Research Procedia. 15. 495-
512. 10.1016/j.trpro.2016.06.042.
Intelligent transit stimulates the development of nation’s [19] Nandury, Satyanarayana & Ara Begum, Beneyaz. (2016). Strategies to handle
infrastructure and improves the sense of happiness of residents. big data for traffic management in smart cities. 356-364. 10.1109/
Through this article, developers discussed significant growth along ICACCI.2016.7732072.
[20] A. Vankudre, M. Bhosale, Detection of Real-Time Traffic through Twitter
with data Analytics and also the recognizing with ITS. Effective Stream Analysis, International Research Journal of Advanced Engineering and
structure for the implementation as concerns ITS data analytics Science ISSN (Online) 2455–9024 (2017) 2.
9
A. Mohandu and M. Kubendiran Materials Today: Proceedings xxx (xxxx) xxx
[21] M.M. Rathore, A. Paul, W.-H. Hong, H. Seo, I. Awan, S. Saeed, Exploiting IoT and [32] A.I. Torre-Bastida, J. Del Ser, I. Laña, M. Ilardia, M.N. Bilbao, S. Campos-
Big Data Analytics: Defining Smart Digital City using Real-Time Urban Data, Cordobés, Big Data for Transportation and Mobility: Recent Advances, Trends
Sustainable Cities and Society. 40 (2017), https://doi.org/10.1016/j. and Challenges, IET Intel. Transport Syst. 12 (8) (2018) 742–755, https://doi.
scs.2017.12.022. org/10.1049/itr2.v12.810.1049/iet-its.2018.5188.
[22] Amini, Sasan & Gerostathopoulos, Ilias & Prehofer, Christian. (2017). Big data [33] J. Hopkins, P. Hawking, Big Data Analytics and IoT in logistics: a case study, The
analytics architecture for real-time traffic control. 710-715. 10.1109/ International Journal of Logistics Management. 29 (2) (2018) 575–591, https://
MTITS.2017.8005605. doi.org/10.1108/IJLM-05-2017-0109.
[23] M. Marjani, F. Nasaruddin, A. Gani, A. Karim, I. Hashem, A. Siddiqa, I. Yaqoob, [34] R. Iqbal, F. Doctor, B. More, S. Mahmud, U. Yousuf, Big data analytics:
Big IoT Data Analytics: Architecture, Opportunities, and Open Research Computational intelligence techniques and application areas, Technol.
Challenges, IEEE Access 5 (2017), https://doi.org/10.1109/ Forecast. Soc. Chang. 153 (2018), https://doi.org/10.1016/j.
access.2017.2689040. techfore.2018.03.024.
[24] Shukla, Sumit & Champaneria, Tushar. (2017). Survey of various data [35] Zhu, Li & Yu, Fei & Wang, Yige & Ning, Bin & Tang, Tao. (2018). Big Data
collection ways for smart transportation domain of smart city. 681-685. Analytics in Intelligent Transportation Systems: A Survey. IEEE Transactions
10.1109/I-SMAC.2017.8058265. on Intelligent Transportation Systems. PP. 1-16. 10.1109/TITS.2018.2815678.
[25] E. Ahmed, I. Yaqoob, I.A.T. Hashem, I. Khan, A.I.A. Ahmed, M. Imran, A.V. [36] S. Dabiri, K. Heaslip, Transport-domain applications of widely used data
Vasilakos, The role of big data analytics in Internet of Things, Comput. Netw. sources in the smart transportation: A survey, Researchgate. (2018) 1–53.
129 (2017) 459–471, https://doi.org/10.1016/j.comnet.2017.06.013. [37] S. Shadroo, A.M. Rahmani, Systematic Survey of Big Data and Data Mining in
[26] Liu, Yang. (2018). Big Data Technology and Its Analysis of Application in Urban Internet of Things, Comput. Netw. 139 (2018) 19–47, https://doi.org/10.1016/
Intelligent Transportation System. 17-19. 10.1109/ICITBS.2018.00012. j.comnet.2018.04.001.
[27] Tomar, Pradeep & Kaur, Gurjit & Singh, Prabhjot. (2018). A Prototype of IoT- [38] Ana Isabel Torre-Bastida1, Javier Del Ser et al. Big Data for transportation and
Based Real Time Smart Street Parking System for Smart Cities. 10.1007/978-3- mobility: recent advances, trends and challenges. IET Journals. June (2018),
319-60435-0_10. Bizana, Spain, pp. 742-755
[28] E. Siow, T. Tiropanis, W. Hall, Analytics for the Internet of Things: A Survey, [39] Neelam, Sahil & Sood, Sandeep. (2019). Smart Vehicular Traffic Management:
ACM Comput. Surv. 51 (4) (2018) 1–36, https://doi.org/10.1145/3204947. An Edge Cloud centric IoT based Framework. 10.1016/j.iot.2019.100140.
[29] M. Ge, H. Bangui, B. Buhnova, Big Data for Internet of Things: A Survey, Future [40] M. Babar, F. Arif, Real-time data processing scheme using big data analytics in
Generation Computer Systems. 87 (2018) 601–614, https://doi.org/10.1016/ internet of things based smart transportation environment, J. Ambient Intell.
j.future.2018.04.053. Hum. Comput. 10 (10) (2019) 4167–4177, https://doi.org/10.1007/s12652-
[30] M. Gohar, M. Muzammal, A. Ur Rahman, SMART TSS: Defining Transportation 018-0820-5.
System Behavior using Big Data Analytics in Smart Cities, Sustainable Cities [41] Amin, Mohammed & Hadouaj, Sameh & Darwish, Tasneem. (2019). Big Data
and Society. 41 (2018) 114–119, https://doi.org/10.1016/j.scs.2018.05.008. Role in Improving Intelligent Transportation Systems Safety: A Survey.
[31] M.A. Javed, S. Zeadally, Data Analytics for Cooperative Intelligent Transport 10.1007/978-3-030-12839-5_18.
Systems, Veh. Commun. 15 (2018), https://doi.org/10.1016/j.
vehcom.2018.10.004.
10