Big Data Analysis in Urban Planning

Download Full-Text PDF Cite this Publication

Text Only Version

Big Data Analysis in Urban Planning

Prit Modi

Dept. of Computer Engineering, Institute of Technology,

Nirma University, Ahmedabad, India

Himani Shah

Shaival Shah

Dept. of Computer Engineering, Institute of Technology,

Nirma University, Ahmedabad, India

Dept. of Computer Engineering, Institute of Technology, Nirma University,

Ahmedabad, India

AbstractBig Data Analytics can be utilized in many domains for analysis of large amounts of data. There has been a vast growth in population in various cities and a rapid rise is required for developing the infrastructure for meeting the increasing needs. In this paper, we analyze various Internet technologies to combine and share a common communication medium. Using this particular knowledgee, an IoT based smart system comprising of all types of sensor deployments which include smart home sensors, vehicular networks, smart parking systems, weather and water sensors and surveillance objects has been proposed which we have reviewed. A four-tier approach has been identified and implemented using Hadoop framework and Spark, volt or S4 for real time processing. Also, the offline historical data needs to be fed into the MapReduce framework of Hadoop. The various IoT datasets generated by all kinds of sensors mentioned above are also used for the analytics and study. The efficiency of the system over here has been calculated using the throughput time and also the processing time.

KeywordsBig data, urban planning, IoT (Internet of Things), Hadoop, Spark

  1. INTRODUCTION

    There is an enormous number of objects being connected to the Internet worldwide at a massive rate. It is supposed to surpass the 50 billion mark, hence resulting in the enrichment of the digital world. The applications of IoT are vast increasing and people are adopting all kinds of IoT devices in their day-to-day lives for their own enrichment. There are a lot of vast domains where IoT has been deployed, like health care, automation, transportation and emergency services etc. IoT transforms regular objects from being normal to its ubiquitous and pervasive computing, embedded devices, communication technologies, sensor, internet protocols in order to change human life.[2]

    The IoT is viewed as a huge prospect for the universe of the Internet. Subsequently, this leads us to the idea of keen homes where distinctive electronic machines are interconnected with one another and accomplish high caliber two-way intelligent mixed media administrations. In such a system where an enormous number of gadgets are interacting with one another, a monstrous volume of information (called Big Data) is produced. To enhance savvy home innovation, the investigation of Big Data could prove an imperative job in the promotion advancement of Information and Communications Technologies (ICTs). This kind of Big Data investigation gives a superior understanding and valuable data about the

    future as well as about arranging and advancement, along these lines giving us knowledge into Big Data.[3]

  2. LITERATURE REVIEW AND RELATED WORKS

    Big Data is turning into a concentrate these days in Intelligent Transportation Systems (ITS). A few contextual investigations in ITS like street traffic systems, flow, public transportation systems, rail transportation the executives and control and resources the board. The reason for ITS is to give better administrations to drivers and riders in transportation Systems. Big Data investigation has settled three issues: 1) Data stockpiling, 2) Data examination and 3) Data the executives. Huge Data stages, for example, Apache Hadoop and Spark are fit to preparing gigantic measures of Data.[1]. In view of the necessities of the shrewd city and urban arranging a 4-level design to break down IoT and Big Data which helps in setting up brilliant urban communities.[2] Here, the generator results are reported and utilized for some applications, for example, flood location, security, and city arranging. This traffic information was then changed over into arrangement records to have the option to process them utilizing Hadoop.[2]

    Urban travel time estimation has been solved given the following challenges: effective modeling approach, data variety and traffic variation patterns.[6] By providing proper travel time information the customers can decide which particular path to choose, mode of travel and when to travel. Here, we consider road segments in a road network under different traffic conditions, in different time slots.[6]

    Paper proposed gives us an easy Real-Time shrewd traffic Management System to give better administration by sending traffic markers to refresh the traffic subtleties immediately. The Real time gushing information is sent for Big Data examination. A versatile application is created as UI to investigate the thickness of traffic at different places and gives an elective method to dealing with the traffic.[5]

    A solution using a combination of dynamic service and the Big Data is described in more detail. This dynamic service is influenced by many parameters and one of them can be the movement of the population. The movement of the population can be monitored through the Big Data.[7] Attempts to explore for a new method of location selection for sculpture planning based on GIS and in combination with the practical case of Taizhou Urban Sculpture Planning. It demonstrates the viability of the technique looked into for

    area determination in figure arranging. It is as yet important for the creator to conduct further research [4].

    This project is implemented by layers, Data Creation, Communication Mediums, Data Analysis and Storage Applications using real time analysis. A dual-node Hadoop Cluster for Data Analysis from sensor data has been implemented [3].

    Map Reduce is used on filtered dataset. Then using these results people will be notified about the traffic congestion, air pollution, water consumption, parking space available at different places.[9] Enormous information preparation for systematic route formation is a huge problem nowadays. In order to solve this, a proposal of a smart city framework, IoT utilizing the investigation of Large Information Sensors. They include keen home, vehicular, climate and in addition water sensors.[8]

    With the challenges that are faced in urban areas, problems in underground pipelines' management pattern is faced, and so the concept of big data thinking is used. Using the knowledge of big data and determining the use of various resources at different places, different underground pipelines can be placed around the city [10].

  3. MOTIVATION

    As referenced before, shrewd urban communities become more brilliant due to the improved idea of computerized innovation, where the keen city is outfitted with various electronic gear used by different applications, for example, road cameras for the observation framework, sensors for the transportation framework, etc. Notwithstanding, there are likewise activities that use items to give distinctive esteem included administrations, such as Google road, the worldwide situating framework (GPS). The various features in a heterogeneous environment and so on, various questions arise like how to tackle uncertainty due to real-time and offline dynamics, how to make current objects smarter and to reduce the data collection cost. The justification behind our aim is to enhance the huge organization of ICT assets in building up the whole framework. Hence, we realize that the headway of late innovation in the installed framework delineates the patterns of ICT. Along these lines, a frmework is required that could in-robust the majority of the ongoing improvements in the field of ICT, because of which amazing development can be seen sooner rather than later. The structure of this framework requires the majority of the capacities of detecting the earth and examining the detecting data. Along these lines, different constant activities could happen because of these innovative assets.

  4. IOT BASED URBAN PLANNING AND SMART CITIES DEVELOPMENT

    After the text edit has been completed, the paper is ready for the template. Duplicate the template file by using the Save As command, and use the naming convention prescribed by your conference for the name of your paper. In this newly created file, highlight all of the contents and import your prepared text file. You are now ready to style your paper; use the scroll down window on the left of the MS Word Formatting toolbar.

    1. IoT based Smart City and Urban Planning

      IoT is basically the connection of various sensors and devices to gather useful information. So as to accomplish our objective, we conveyed numerous sensors at better places to gather and break down information for better use. A definitive objective is to accomplish savvy homes, stopping, climate and water frameworks, vehicular traffic, ecological contamination and reconnaissance frameworks.[7]

      In a smart home, the house is consistently observed by sending information created from the sensors estimating the smoke and temperature, water utilization of houses in various regions of the city. Observing contamination helps in the human services of the natives and alarms them when the contamination increments over a specific edge.

      Smart Parking helps by following the vehicles traveling every which way out of various vehicle leaving zones. In this way, smart vehicle leaving can be structured, thinking about the quantity of vehicles in a district, or new vehicle leaving can be created where there are more autos generally.

      Climate and water data additionally build the effectiveness of the smart city by giving climate related information, for example temperature, downpour, stickiness, weight, wind speed and water levels in streams, lakes, dams, and different stores. The majority of this data is gathered by setting the sensors in water stores and other open spots. We can likewise anticipate data relating to water repositories ahead of time to meet the residents' water needs.

      Vehicular traffic data is the most noteworthy wellspring of information in a shrewd city. Through this kind of information source and with helpful continuous examination, residents and the administration will profit enormously. Natives can decide to what extent it will take to achieve a goal dependent on the present power of traffic and the normal speed of the vehicles. We get the area of every vehicle and the quantity of vehicles between two sets of sensors set at different areas in the city. Additionally, if any mishap has occurred, the windshield will be harmed and the sensor will send a caution to the police, traffic specialists, and the clinic. A city can never be shrewd with unfortunate residents. Thus, while structuring the smart city, we planned a different module to get ecological information that incorporates data on gases, for example, specific metals, carbon monoxide, sulfur dioxide and ozone. These gases are hazardous to human well being and can cause liver issue and coronary illness. Individuals ought not to go outside when these gases are at abnormal states in the earth, particularly kids, the older, the individuals who are participating in physical exercise, or debilitated. [5]

    2. Architecture and Implementation of Big Data

    Tier I. Bottom tier: This layer handles data generation through IoT sources. As there are many IoT sensors, huge data is produced. In addition, different information has security, protection, and quality prerequisites. Furthermore, with respect to sensor information. Various filtration methods are applied at this layer, which channels the pointless metadata, and repeated information are likewise disposed of. Tier-II. Intermediate tier-I: This level is in charge of the correspondence among sensors and from sensors to transfer hubs through ZigBee innovation, and depends on GW or base station and after that on the Internet utilizing different

    communication advancements, for example, Wi-Fi, WiMAX, LTE,3G, and so on. Ethernet is utilized between different analysis servers. [8]

    Tier-III. Intermediate tier-II: This layer is the principal layer of the whole diagnostic framework and is in charge of information handling. We need continuous examination for the brilliant framework; in this manner, we need an outsider constant apparatus to consolidate with Hadoop to give ongoing usage. To give ongoing execution, Strom, Spark, or VoltDb could likewise be utilized. Nonetheless, for framework assessment, we executed the framework with Spark. At the lower layer of Hadoop, a similar structure of MapReduce and HDFS is utilized. With this framework, we can likewise utilize HIVE, HBASE, and SQL for overseeing Database (in-memory or Offline) to store chronicled data. For urban arranging since ongoing outcomes were irrelevant, we utilized Hadoop with MapReduce programming.

  5. ANALYZING URBAN DATA

    To properly analyze and do the relevant study we need to have a large number of IoT datasets. Here, we need to understand how a particular setup matters, deployment of different sensors and also this methodology can be applied to real-time data as well.

    1. Dataset Description

      An enormous IoT-created dataset from different capable assets have been taken. The datasets incorporate:

      1) The information of floods happening everywhere throughout the world, 2) shrewd home temperatures, counting the water utilization of each house, etc. 3) vehicular datasets, including all subtleties of the vehicles traveling between many source and goal focuses at different parts in the city, 4)stopping place datasets, including the current status of number of vehicles in the leaving region,

      5)contamination datasets, including different gases and commotion pollution, 6) online networking datasets, for example, Twitter, including every day tweet record, 7) climate datasets, including continuous estimations of temperature, stickiness, downpour outside and inside the home, 8) other information normal city datasets, for example, social and library occasions. Complete dataset subtleties, including dataset estimate.[4]

      Fig. 1. Implementation Model

    2. Analysis of Various Datasets

      1. Vehicular Dataset Analysis

        The quantity of vehicles in a specific region plays a vital job in the public arena. For example, during on hours, the traffic power on specific streets is higher than during off times. Likewise, the street the board framework can be affected by the quantity of vehicles at a specific time and on a specific street.

        From the above IoT-based system traffic investigation, we can foresee the evaluated time to achieve one point from the other point. The keen city breaks down vehicular traffic information progressively and encourages discovering how much time it will take to achieve a goal by following elective courses contingent upon the present force of the traffic. Its master videos refreshed data with respect to travel so individuals can plan to achieve the goal by following the most advantageous course. It also helps the government in properly laying out the road structure of the city. Whatever happens on the road, for example accidents, potholes etc. can all be taken into account and accordingly the traffic management will take place. Hence, the savvy city will help in averting the entire traffic in another route and hence help us in avoiding obstacles.

        Fig. 2. Intensity of traffic on Madrid highways at different locations

      2. Parking Lot Data

        As per this investigation, clients are updated about free parking continuously and would thus be able to spare fuel without manually scanning. Moreover, this likewise proves to be beneficial for sellers in the city who are not making much profit. For the most part, residents like to go out to shop where it is less clogged and where parking is promptly accessible, bringing about more noteworthy benefit for merchants in such territories. The examination of parking likewise gives guidance to legislative experts in urban planning to assemble all the more parking territories close places with a higher volume of customers.

      3. Environmental Data Analysis for Pollution

    For the motivations behind examination and to keep the gas esteems inside a cut off, the counts of gas esteems are marginally changed. Be that as it may, this does not influence the analysis of the truth of impact of gases. The estimations of vehicle carbon monoxide, nitrogen dioxide, sulphur dioxide, particulate matter, and ozone level gas esteems are determined as follows:

    • Initially doled out an incentive somewhere in the range of 25 and 100. Each 5 min, the quality is refreshed as follows:

    • If the esteem was underneath 20 preceding, it would now be the last value + some random number somewhere in the range of 1 and 10.

    • If the esteem was higher than 210, it would now be the

      last arbitrary whole number somewhere in the range of 1 and 10.

    • Otherwise, the esteem is the last value an arbitrary integer somewhere in the range of 5 and 5.

    These gases are hazardous when their qualities are more noteworthy. The most extreme qualities for all gases are depicted as ozone esteem now and again 7090, particulate matter qualities at time 185215 and more than 245, nitrogen dioxide toward the beginning and end of the time, and carbon monoxide at 90115, are for the most part risky for well- being. Along these lines, youngsters ought not be permitted to spend time outside during these occasions.

  6. SYSTEM IMPLEMENTATION IN REAL TIME

    This has been implemented on two main modules which also has many sub-areas included.

    1. Implementation on Smart City

      The information source remains as depicted in Fig 3. Each facility of the brilliant city is executed as an isolated class or sub-module that takes information from different sources. Traffic data estimations take information from vehicular traffic and parking. The security module takes information from

      Fig. 3. .Implementation of Smart City

      surveillance, smart homes, and vehicular traffic for the situation that the administration needs to screen for stolen vehicles. The flood and water management modules take water use information from smart home information for downpour and ice storms additionally foresee floods continuously. Also, energy utilization management takes power and gas information from savvy homes just as dam and water-related information. This module oversees and spares additional energy that is not being utilized by homes. It likewise disseminates the energy to different territories as indicated by the need. At last, health management module settles on choice with respect to contamination information. Residents have restricted access to these modules, however the administration has full access to them.

    2. Implementation on Urban Planning

    Urban arranging framework execution is led at three dimensions, i.e. the Physical, Halfway and Upper level. The physical dimension is known as the Storage Level and depends on the Hadoop HDFS framework. Every single authentic datum are stored at the physical dimension. The moderate dimension is the second dimension, which is additionally called the Processing Level. All preparing is performed at this level utilizing the information stored on the physical level. At this level, measurable estimation, calculation, diagram examination is performed. The upper level also known as the Decision Level. Here, various decisions regarding the urban planning and various modules for each type of planning are being made at this level.

  7. EVALUATION

    The proposed algorithm was implemented using Single Hadoop single node setup. Since the framework depends on Big Data examination, it was assessed regarding effectiveness and reaction time. Framework execution was estimated for different measured datasets considering the handling time (in milliseconds also including the throughput (Mbps). The processing time results are appeared in Fig. 4, and the throughput investigation results appear in Fig. 5.

    Fig. 4. Processing times of various sized vehicular datasets

    Fig. 5. Throughput of datasets depending on the size of data

    It is evident from the chart that when information size increases, the processing time increases accordingly. The processing can always be tested at higher sized dataset. Besides, when we dissected the throughput relating to the information measure, we identified that the throughput was additionally straightforwardly corresponding to data size due to the parallel handling nature of the Hadoop framework. This is the significant accomplishments of the system, as when there is an expansion in information estimate, the throughput is likewise expanded. The throughput of the system with respect to the number of sensors is shown in Fig. 6.

    Fig. 6. Throughput of the system by increasing the number of sensors per record for 1 GB of data

  8. CONCLUSION

Smart cities and urban planning play a very vital role in national development. In this paper by using IoT generated big data analysis. This includes four layers of having functionalities like collection, aggregation, communication, processing, and interpretation. The system is developed using Hadoop and real time analysis is carried on datasets vehicular networks, smart parking, smart home, weather and pollution. The proposed system is evaluated on the basis of efficiency which has been increasing almost linearly with the amount of data-set record.

REFERENCES

  1. L. Zhu, F. R. Yu, Y. Wang, B. Ning, and T. Tang, Big Data Analytics in Intelligent Transportation Systems: A Survey, IEEE Trans. Intell. Transport. Syst., vol. 20, no. 1, pp. 383398, Jan. 2019.

  2. M. M. Rathore, A. Ahmad, A. Paul, and S. Rho, Urban planning and building smart cities based on the Internet of Things using Big Data analytics, Computer Networks, vol. 101, pp. 6380, Jun. 2016.

  3. B. N. Silva et al., Exploiting Big Data Analytics for Urban Planning and Smart City Performance Improvement, in 2018 12th International Conference on Signal Processing and Communication Systems (ICSPCS), Cairns, Australia, 2018, pp. 14.

  4. S. Xu, C. Gong, and Z. Tong, Research on Location Selection of Urban Sculpture Based on Big Data Case Study of Taizhou, in 2018 26th International Conference on Geoinformatics, Kunming, 2018, pp. 14.

  5. P. Rizwan, K. Suresh, and M. R. Babu, Real-time smart traffic management system for smart cities by using Internet of Things and big data, in 2016 International Conference on Emerging Technological Trends (ICETT), Kollam, India, 2016, pp. 17.

  6. K. Tang, S. Chen and Z. Liu, "Citywide Spatial-Temporal Travel Time Estimation Using Big and Sparse Trajectories," in IEEE Transactions on Intelligent Transportation Systems, vol. 19, no. 12, pp. 4023-4034,

    Dec. 2018.doi:10.1109/TITS.2018.2803085

  7. P. Horaovský, V. Novotný and M. Svítek, "Data-driven management of dynamic public transport," 2018 Smart City Symposium Prague (SCSP), Prague, 2018, pp. 1-5. doi: 10.1109/SCSP.2018.8402665

  8. P. Yadav and S. Vishwakarma, "Application of Internet of Things and Big Data towards a Smart City," 2018 3rd International Conference On Internet of Things: Smart Innovation and Usages (IoT-SIU), Bhimtal, 2018, pp. 1-5. doi: 10.1109/IoT-SIU.2018.8519920

  9. A. Imawan and J. Kwon, "A timeline visualization system for road traffic big data," 2015 IEEEInternational Conference on Big Data (Big Data), Santa Clara, CA, 2015, pp. 2928-2929.

  10. H. Wan and D. Kong, "Research on Innovation of Management Concept of Urban Underground Pipelines Based on Big-Knowledge- Thinking in the Big Data Era," 2017 IEEE International Conference on Big Knowledge (ICBK), Hefei, 2017, pp. 220-223

Leave a Reply

Your email address will not be published. Required fields are marked *