Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

200712103

687 views

Published on

  • Be the first to comment

200712103

  1. 1. REMOTE SENSING AND GIS BASED PAVEMENT PERFORMANCE PREDICTION MODEL USING ARTIFICIAL NEURAL NETWORK <ul><li>UNDER THE GUIDENCE, </li></ul><ul><li>Dr.C.UDHAYA KUMAR </li></ul><ul><li>ASST. PROFESSOR </li></ul><ul><li>IRS, ANNA UNIVERSITY </li></ul><ul><li>BY, </li></ul><ul><li>DEVI PRIYADARISINI.K </li></ul><ul><li>ROLL NO: 200712103 </li></ul><ul><li>M.E. GEO INFORMATICS </li></ul><ul><li>IRS, ANNA UNIVERSITY. </li></ul>
  2. 2. INTRODUCTION <ul><li>Pavement surface are a major component of Infrastructure. </li></ul><ul><li>The existing route system has become structurally inadequate. </li></ul><ul><li>GIS can add tremendous functionality to a pavement management condition program. </li></ul>
  3. 3. Objectives <ul><li>To map the present condition of the pavement. </li></ul><ul><li>To develop pavement performance predicition model. </li></ul><ul><li>To incorporate the model with GIS and create graphical outputs. </li></ul><ul><li>To validate the model. </li></ul><ul><li>To predict and map future condition. </li></ul>
  4. 4. Scope of Work <ul><li>To monitor and maintain the major infrastructure asset ,highway. </li></ul><ul><li>The knowledge of future pavement performance is essential to PMS. </li></ul><ul><li>GIS environment to support the pavement management decision making using several application. </li></ul><ul><li>Which would help in predict the pavement performance accurately . </li></ul>
  5. 5. Study Area <ul><li>Location : Chennai </li></ul><ul><li>Stretches : Sadar Patel Road – 3000 m </li></ul>
  6. 7. Data to be Used <ul><li>Non – Spatial Data </li></ul><ul><li>Spatial Data </li></ul><ul><li>Structure of the pavement. </li></ul><ul><li>Surface Parameter </li></ul><ul><li>Cracks </li></ul><ul><li>Potholes </li></ul><ul><li>Rut Depth </li></ul><ul><li>Skid Residence </li></ul><ul><li>Soil Characteristics </li></ul><ul><li>Slope </li></ul><ul><li>Quick bird – 0.6m </li></ul><ul><li>SOI Toposheets : 66C 4 , C8 </li></ul>
  7. 9. Calculation of PSI <ul><li>The PSI value is calculated by using the following equation : </li></ul><ul><li>PSI = 20.715 – 6.676 *log (R) – 0.0283 * D </li></ul><ul><li>Where, </li></ul><ul><li>R – Unevenness Index of the pavement surface </li></ul><ul><li> D – Total Surface distress. </li></ul>
  8. 10. Pavement Condition Scale
  9. 11. Present PSI Scale Along S.P Road
  10. 12. Present Condition Map
  11. 13. Development of ANN Structure
  12. 14. Comparison of ANN architecture of the model
  13. 15. Comparison of ANN and PSI for PPP for testing set
  14. 16. Comparison of ANN and PSI for PPP for testing set
  15. 17. Predicted PSI value along SP Road
  16. 18. Future condition map
  17. 19. Pavement Condition Distribution Graph
  18. 20. Future Condition Distribution
  19. 21. Conclusion <ul><li>Development and use of PMS using Pavement attribute database in GIS environment. </li></ul><ul><li>Different types of operation can be performed. </li></ul><ul><li>The developed ANN model can be used for several Pavement Management decision. </li></ul><ul><li>The ANN model developed gives out better results than the PPP to the AASHTO panel data. </li></ul><ul><li>Effective decision making in Pavement Management System. </li></ul><ul><li>Analysis and budget allocation for rehabilitation purpose. </li></ul>
  20. 22. Future Recommendation <ul><li>Applying the PSI and ANN model concept criterion to setup maintenance priorities, maintenance cost and pavement management programs. </li></ul><ul><li>Adapting GPS and GIS based vision systems for the purpose of distresses data collection and measurements. </li></ul>
  21. 23. References <ul><li>Barry White and Alex Rocie, “GIS and Pavement Management”, Transportation Engineering ASCE. </li></ul><ul><li>San Diego, “Development of GIS based Illinois Highway pavement management”, ESRI user conference. </li></ul><ul><li>Deva Pratap, Kiran Kumar, “Highway Information System and Management using GIS”, Gis development. </li></ul><ul><li>Michael T. Me Nernery and Thomas Row, “GIS need assessment for TxDOT pavement information system”, US Dept. of Transport. </li></ul><ul><li>Andres L.Bako, Zoltan Hervath, “Decision Supporting Model for Highway Maintaince”, Journal of Infrastructure System, ASCE. </li></ul><ul><li>Gerardo W.Flintish , Randy Dymond and Jhon Collua, “Pavement Management Application Using GIS”, NCHRP 2005, pp 1-25. </li></ul><ul><li>Serdal Terzai, “Modeling the pavement serviceability ratio of flexible highway pavement by ANN”, ELSEVIER, Construction and Building Materials 17 (2007) 577–582 </li></ul><ul><li>  </li></ul>
  22. 24. Contd… <ul><li>Mohammed Taleb Obaidat , Sharaf A. Al-kheder, “Integration of geographic information systems and computer vision systems for pavement distress classification”, ELSEVIER, Construction and Building Materials 20 (2006) 657–672 </li></ul><ul><li>J.A.Prozji, S.M.Madanat, “Developing of Pavement Performance models combing Experimental field data”, Journal of Infrastructure System, ASCE. </li></ul><ul><li>Samer Madana, Jorge A.Prozzi and Michael Han, “Effect of performance model accuracy on optimal pavement design”, Journal of Infrastructure System, ASCE 2006 August. </li></ul><ul><li>Abul Hamid Modh Isa, Law Tick Hwa, Dadarcy Mohamed Ma’soen , “Pavement performance models for federal roads&quot;, Journal of Infrastructure System, ASCE, 2007, April </li></ul><ul><li>Bosurgi G., “Artificial Neural Networks for Predicting Road Pavement Conditions”, 4 th International SIIV Congress – PALERMO (ITALY), 12-14 September 2007. </li></ul>

×