Disaggregate Traffic Forecasting and Diversion analysis for Inter-Regional Traffic Corridor
- 1. P.G. section in Transportation Engineering & Planning DEPARTMENT OF CIVIL ENGINEERING SARDAR VALLABHBHAI NATIONAL INSTITUTE OF TECHNOLOGY SURAT, INDIA DISSERTATION (CE-864) “DISAGGREGATE TRAFFIC FORECASTING AND DIVERSION ANALYSIS FOR INTER-REGIONAL TRAFFIC CORRIDOR USING ECONOMETRIC APPROACH” Presented by : Rahul Deshwal (P17TP003) Guided by: Prof. G. J. Joshi Dr. S. S. Arkatkar 1
- 2. CONTENTS I. Introduction II. Need for Study III. Literature Review IV. Objectives and Scope of Work V. Methodology VI. Study Area VII.Data Collection and Analysis VIII.Employment Forecasting IX. Traffic Prediction for NH-48 X. Diversion Analysis XI. Results and Discussion XII.References 2
- 3. Introduction Planning is the process of thinking about the activities required to achieve a desired goal. Planning of the region is done in two aspects: Economic planning, and Physical planning. Transportation system plays an important role in economic, industrial, social and cultural development for the region. Hence, Regional planning of transport infrastructure for the region's connectivity and circuity with the other regions becomes crucial. Regional transportation planning provides integration of transportation plan with land use plan of the region. Road transport in India is the dominant mode of transport 65% of freight movement, and 85% of passenger traffic. The National Highways network of 115,435 km contributes to about 2.7% of the country’s road network and carries about 40% of total traffic. 3
- 4. Need of Study The Mumbai–Ahmedabad Corridor in the western part of the country is one of the important transport corridors of the country. Industries like, textile industry, gems and jewelleries, petrochemical & fertilizer and other industrial complexes have been established along this corridor. According to EIA report submitted by ICT Pvt. Ltd., The average journey speed on Mumbai-Vadodara section of NH-48 was found 50-60 km/hr whereas a NH is designed for the speed of 100 km/hr. NH-48, 6-lane highway, was carrying traffic in the range of 50,000 to 80,000 PCU per day in 2013 and reached LOS E in 2015 itself. Report has growth factor and elasticity based aggregate traffic forecast which undermines the individual growth patterns of different vehicle class. 4
- 5. Author Title Review and Findings Stephen Fretz, Christoph Gorgas (2013) Regional economic effects of transport infrastructure expansion: evidence from the Swiss highway network. The main aim of the study was to quantitatively assess the impact of improved regional accessibility. Variables such as GDP, unemployment rate and industrial structure. They concluded that improved transport infrastructure network shows improvement in intangible benefits at regional level, and also labour market play an important role for infrastructure expansion to generate benefits in terms of income. Ying Jin, Ian Williams (2000) A new regional economic model for European transport corridor studies. The author reported the development of new operational model for assessing the socio-economic impact of strategic transport initiatives. They have described the likely use of regional economic model for transport. They linked transport demand forecast with regional economy via regional economic model and assessed the impact of transport in context of that regional economy. Literature Review: Regional Economy5
- 6. Author Title Review and Findings Jim Yoo Kim, Jung Hoon Han. (2015) Straw effects of new highway construction on local population and employment growth. The study was carried out to examine the rising concern of straw effects of new highway using difference-in- difference model to measure effects of new highway on local population and employment growth. The results revealed that no evidence of the straw effects were found and not only the new highways contribute to increasing population and employment in lagging areas but also improve the accessibility of existing highway. Uwe Blien, Tassinopoulos (2001) Forecasting regional employment The authors described forecasting of employment was based on entropy optimizing procedure, a newly developed for the estimation of matrices from heterogeneous information. Simple linear regression using first differences between years was developed for estimating the trend of employment. They found employment forecasting becomes necessary as the change in employment causes change in vehicle ownership. Literature Review: Employment Forecasting6
- 7. Author Title Review and Findings Kartikeya Jha, Nishita Sinha, S.S. Arkatkar (2012) Modelling Growth Trend & Forecasting Techniques for Vehicular Population in India. Trend Line Analysis of last 25 years was carried out. Econometric regression models were also developed using variables such as population and per capita income. Time Series Analysis was also done using the Box and Jenkins Methodology. They found ARIMA models of Box and Jenkins methodology perform better than trend line analysis and econometric regression analysis. Shabana Thabassum (2013) Impact of state-wise vehicle contribution on traffic growth rates for National Highways The impact of different state vehicular growth on the state highway. Socio-economic variable such a population, NSDP and per capita income (PCI) were used to determine the elasticity of transport demand and develop regression models. Traffic growth rate for different categories of vehicle was determined, and forecasted traffic was used for financial analysis and pavement design. Indian Road Congress (2015) IRC: 108-2015 Guidelines for Traffic Forecast on Highways It provides methodology for traffic forecasting process. Literature Review: Traffic Forecasting7
- 8. 8 Stage – 3 Forecasting of Travel Stage – 2 Analysis of Base Year Travel Stage – 1 Preparatory Works Study of the Base Year Transport Network Map Reconnaissance Survey Identification of the PIA and TAZs Identification of the Homogenous Traffic Sections Identification of Primary Surveys and their Locations Primary Traffic Studies Base Year Traffic Analysis Normal Traffic Forecast Identification of Secondary Data and their Sources Collection of Secondary Data Estimation of Growth Factor for Normal Traffic Developmental Traffic Forecast Total Traffic Forecast Horizon year Transport Network Generated Traffic Forecast Source:- IRC:108-2015 Fig. A.1 Traffic Forecasting Process
- 9. Contd. Traffic Growth Forecast 9 AADT Normal Traffic Forcast • Past Traffic Trend • Past trend of Vehicle Population • Elasticity of Transport Demand • Time Series (SMA, DMA, SES, ARMA, ARIMA, etc.) Developmental Traffic Forecast • New township • Industrial unit or SEZ Generated Traffic Forecast • Diverted Traffic • Induced Traffic
- 10. Objectives The main objectives of this research work are as follows: To analyse historical inter-regional traffic flow pattern of NH 48 on various temporal scale To analyse effect of Regional/National Economic growth on traffic flow of NH 48 To develop appropriate Time Series model for Traffic Forecast and consequent LOS pattern To develop models for Diversion Forecast for alternate highway. 10 Scope of Work Historical traffic data of toll booths in region Economic growth indicators GDP and GSDP Impact of the industrial growth of the study region (South Gujarat) on the National Highway 48. User response survey for diversion analysis Viability of the proposed Vadodara-Mumbai Expressway in terms of travel demand.
- 11. 11 Primary Data Secondary Data Methodology
- 12. The National Highway passes from Major Cities:- • Delhi • Gurugram • Jaipur • Ajmer • Beawar • Udaipur • Ahmedabad • Vadodara • Surat • Mumbai 12 NH-48 from Delhi to Mumbai
- 13. 13 STUDY REGION
- 14. DISTRICTS AND CITIES IN STUDY REGION14 Study Region consist of 5 Districts of South Gujarat region: • Vadodara, • Bharuch, • Surat, • Navsari, • Valsad. Source: Draft EIA report submitted by ICT Pvt. Ltd., New Delhi
- 15. 15 CHANGE IN POPULATION OVER DECADE
- 16. 16 CHANGE IN POPULATION DENSITY OVER DECADE
- 17. • Vadodara – 3 • Bharuch – 8 • Surat – 3 • Valsad – 2 Districtwise SEZ's approvals in Region • 18 Private Industrial Park • 42 GIDC Industrial Estate Industrial Park in Region • Dahej • Hazira • Magdalla • Vansi Ports in Region 17 Private Industrial Park Bharuch – 11 Surat – 3 Valsad – 2 Vapi – 2 GIDC Industrial Estate Bharuch – 17 Surat – 14 Valsad – 7 Navsari – 4 SEZ AND INDUSTRIES IN REGION
- 18. NH 53 NH 360 Toll Booth 18 Existing Highway in Study Region
- 19. Phase 1 consist of construction of 274 km road stretch 260.4 km is in state of Gujarat, 5.5 km in Union Territory of Dadra and Nagar Haveli and, 8.1 km in district of Thane in state of Maharashtra Existing NH 48 in the same corridor has stretch of 277 km in Gujarat. • VME is passing through • Vadodara (54.4km), • Bharuch (62.5 km), • Surat (57.3km), • Navsari (37.6km) and • Valsad (48.6km) in the state of Gujarat (260.4km). 19 Vadodara Mumbai Expressway (VME)
- 20. 20 DATA
- 21. DATA COLLECTION AND ANALYSIS21 Primary Data (Questionnaire Survey) Revealed Preference data Vehicle Characteristics Travel Characteristics Stated Preference data Priority Ranking Willingness to Pay Secondary Data Classified monthly traffic data of toll Road Network GDP/GSDP data Industry data RTO data
- 22. SECONDARY DATA22 Government Agency Data Collected Period NHAI – PIU Surat Classified Monthly Traffic Volume (Schedule M) of Karjan toll, Boriach toll and Bhagwada toll 2009-2018 MSME – Development Institute Employment and Investment Data 1985-2011 2006-2015 Directorate of Census Operations, Gujarat Population, Household and Employment 1991, 2001, 2011 Planning Commission GDP and GSDP 2001-2018 RTO, Gujarat Vehicle Registration 2000-2018
- 23. TOLL DATA23
- 24. Districts/Region Vadodara Bharuch Surat Navsari Valsad Region Investment Year 32 22 32 16 19 32 CAGR 24.93% 26.15% 29.34% 38.92% 40.62% 28.85% Employment Year 32 14 32 16 19 32 CAGR 17.16% 36.11% 21.01% 25.87% 27.49% 20.60% 0 500000 1000000 1500000 2000000 2500000 3000000 3500000 4000000 4500000 0 10 20 30 40 CumulativeInvestment(LakhRs.) Year Growth in Investment in Districts of Region Vadodara Bharuch Surat Navsari Valsad Region 0 500000 1000000 1500000 2000000 0 10 20 30 40 CumulativeEmployment Year Growth in Employment in Districts of Region Vadodara Bharuch Surat Navsari Valsad Region EMPLOYMENT AND INVESTMENT DATA24
- 25. EMPLOYMENT MULTIPLIER25 Gujarat Vadodara Bharuch Surat Navsari Valsad Region Industry Emp 3.55 0.21 0.08 0.62 0.02 0.10 1.03 Agricultural Emp 10.17 0.88 0.33 0.44 0.32 0.33 2.30 Basic Emp 13.73 1.10 0.41 1.06 0.34 0.43 3.33 Non-Basic Emp 11.04 0.60 0.22 1.50 0.25 0.32 2.88 Total Emp 24.77 1.69 0.63 2.55 0.59 0.74 6.21 Total Population 60.44 4.17 1.55 6.08 1.33 1.71 14.83 WPR 0.41 0.41 0.40 0.42 0.45 0.44 0.42 Base Ratio 0.80 0.54 0.53 1.42 0.73 0.75 0.86 Emp Multiplier 1.80 1.54 1.53 2.42 1.73 1.75 1.86 Emp (Employment) and Population in Million's WPR – Work Participation Rate and Region is Study Area
- 26. SURVEY LOCATIONS26 Preliminary Survey Locations Primary Survey Locations
- 27. 27 A total of 806 samples were collected in preliminary and primary survey removing samples with inappropriate/incomplete data. 70 Samples were collected in preliminary survey on basis of which 5 locations for primary survey were finalised. 736 Samples were collected from 5 locations in 2 days time during primary survey. PRIMARY DATA
- 28. Traffic Entry and Exit locations Origin Destination Gujarat 64 67 Maharashtra 26 20 0 10 20 30 40 50 60 70 80 Percentage% Movement of Vehicle for Gujarat and Maharashtra E N ENE Sura t Valsad E E E S 0 5 10 15 20 25 30 Percentage Location Distribution of Primary data based on O-D pattern Origin Destination 28
- 29. Travel Pattern and Route Preference 0% 20% 40% 60% 80% 100% Car LCV Bus Truck MAV Percentage% Vehicle Category Vehicle Category Travel Pattern Internal to Internal Internal to External External to Internal External to External Car LCV Bus Truck MAV Total Yes 81 70 38 81 85 81 No 19 30 63 19 15 19 0 10 20 30 40 50 60 70 80 90 Percentage% Vehicle Category Preference for Vadodara Mumbai Expressway Region 1 Region 2 Region 3 Region 4 External External External External I-E & E-I E-E I-I 29
- 30. Descriptive Statistics 45% 12% 20% 23% Travel Pattern (No) Internal to Internal Internal to External External to Internal External to External 16% 21% 24% 39% Travel Pattern (Yes) Internal to Internal Internal to External External to Internal External to External 30
- 31. Priority for Road Users 0 50 100 150 200 Travel Time Travel Cost Distance Road Side Amenities Prioirty for Car 1 2 3 4 0 10 20 30 40 50 60 70 Travel Time Travel Cost Distance Road Side Amenities Prioirty for LCV 1 2 3 4 0 20 40 60 80 100 120 140 Travel Time Travel Cost Distance Road Side Amenities Prioirty for Bus_Truck 1 2 3 4 0 50 100 150 200 250 Travel Time Travel Cost Distance Road Side Amenities Prioirty for MAV 1 2 3 4 31
- 33. 33 FORECASTING
- 34. EMPLOYMENT FORECASTING34 District Model Equation R² RMSE MAE MAPE Vadodara INEMPVA y = 0.396x + 25067 0.964 11481 9513 11% Bharuch INEMPBH y = 0.372x - 359 0.979 4389 3530 8% Surat INEMPST y = 0.512x + 74338 0.984 41060 31713 11% Navsari INEMPNI y = 0.337x + 3915 0.933 2229 1693 12% Valsad INEMPVD y = 0.317x + 7024 0.978 6433 4773 8% Region INEMPRN y = 0.463x + 101376 0.986 54175 41013 10% y = Cumulative Employment and x = Cumulative Investment
- 35. IN-EMP MODELS35 0 50000 100000 150000 200000 250000 300000 0 200000 400000 600000 800000 CumulativeEMPLOYMENT Cumulative INVESTMENT (Lakh Rs.) INEMPVA 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 100000 0 100000 200000 300000 CumulativeEMPLOYMENT Cumulative INVESTMENT (Lakh Rs.) INEMPBH 0 200000 400000 600000 800000 1000000 1200000 1400000 1600000 0 1000000 2000000 3000000 CumulativeEMPLOYMENT Cumulative INVESTMENT (Lakh Rs.) INEMPST 0 5000 10000 15000 20000 25000 30000 0 20000 40000 60000 80000 CumulativeEMPLOYMENT Cumulative INVESTMENT (Lakh Rs.) INEMPNI 0 20000 40000 60000 80000 100000 120000 140000 0 100000 200000 300000 400000 CumulativeEMPLOYMENT Cumulative INVESTMENT (Lakh Rs.) INEMPVD 0 500000 1000000 1500000 2000000 2500000 0 2000000 4000000 6000000 CumulativeEMPLOYMENT Cumulative INVESTMENT (Lakh Rs.) INEMPRN
- 36. TRAFFIC PREDICTION FOR NH-4836 Traffic Prediction by Aggregate Approach Traffic Prediction by Disaggregate Approach PCU Values as per Indo-HCM 2017 Vehicle Type Equivalent Factor Car 1 Light Commercial Vehicle (LCV) 3.2 Bus/Truck 5 Multi Axle Vehicle (MAV) 5.5
- 37. AGGREGATE APPROACH37 Employment and Investment Vehicle Registration GDP and GSDP
- 38. EMPLOYMENT, INVESTMENT & AADT RELATIONSHIP38 Variable Equations for Karjan toll R² RMSE MAE MAPE REmp AADT = 0.013REmp + 83442 0.8583 2028 1879 1.91% RIn AADT = 0.005RIn + 87022 0.8504 2083 1893 1.93% REmp-RIn AADT = 0.295REmp - 0.11RIn 0.9923 8750 7700 7.64% Variable Equation for Choryasi toll R² RMSE MAE MAPE REmp AADT = 0.01REmp + 80271 0.7851 1903 1632 1.75% RIn AADT = 0.004RIn + 83173 0.7293 2136 1870 2.00% REmp-RIn AADT = 0.063REmp - 0.021Rin + 65982 0.9323 1068 871 0.97% Variable Equation for Boriach toll R² RMSE MAE MAPE REmp AADT = 0.022REmp + 51885 0.8177 4046 3195 3.96% RIn AADT = 0.009RIn + 57482 0.8519 3647 3036 3.80% REmp-RIn AADT = 0.189REmp - 0.064RIn 0.9891 8406 7539 9.33% Variable Equation for Bhagwada toll R² RMSE MAE MAPE REmp AADT = 0.019REmp + 52774 0.8852 2661 2035 2.61% RIn AADT = 0.008RIn + 57795 0.9034 2441 2028 2.63% REmp-RIn AADT = 0.193REmp - 0.067RIn 0.9917 7036 6225 7.99%
- 39. CONTD.39 80000 90000 100000 110000 120000 80000 90000 100000 110000 120000 PredictedAADT Observed AADT Predicted vs Observed Karjan AADT REmp Rin REmp-RIn 80000 90000 100000 80000 90000 100000 PredictedAADT Observed AADT Predicted vs Observed Choryasi AADT REmp Rin REmp-RIn 60000 70000 80000 90000 100000 110000 60000 80000 100000 120000 PredictedAADT Observed AADT Predicted vs Observed Boriach AADT REmp Rin REmp-RIn 60000 70000 80000 90000 100000 60000 70000 80000 90000 100000 PredictedAADT Observed AADT Predicted vs Observed Bhagwada AADT REmp Rin REmp-RIn
- 40. VEHICLE REGISTRATION & AADT RELATIONSHIP40 Variable (in Million) Equation (AADT in '000) for Karjan toll R² RMSE MAE MAPE RegTV AADT = 22.04RegTV + 61 0.728 4512 3431 3.24% RegV AADT = 2.076RegV + 69 0.753 4302 3296 3.13% Variable (in Million) Equation (AADT in '000) for Choryasi toll R² RMSE MAE MAPE RegTV AADT = 21.244RegTV + 55 0.849 3002 2398 2.49% RegV AADT = 1.997RegV + 64 0.874 2737 2090 2.18% Variable (in Million) Equation (AADT in '000) for Borich toll R² RMSE MAE MAPE RegTV AADT = 28.005RegTV + 30 0.873 3915 3103 3.61% RegV AADT = 2.529RegV + 42 0.876 3867 2807 3.20% Variable (in Million) Equation (AADT in '000) for Bhagwada toll R² RMSE MAE MAPE RegTV AADT = 26.601RegTV + 29 0.906 3143 2638 3.17% RegV AADT = 2.41RegV + 41 0.915 2990 2574 3.10%
- 41. CONTD.41 90000 95000 100000 105000 110000 90000 95000 100000 105000 110000 PredictedAADT Observed AADT Observed vs Predicted Karjan AADT RegTV RegV 80000 90000 100000 110000 80000 90000 100000 110000 PredictedAADT Observed AADT Observed vs Predicted Choryasi AADT RegTV RegV 60000 70000 80000 90000 100000 110000 60000 80000 100000 PredictedAADT Observed AADT Observed vs Predicted Boriach AADT RegTV RegV 60000 70000 80000 90000 100000 110000 60000 80000 100000 PredictedAADT Observed AADT Observed vs Predicted Bhagwada AADT RegTV RegV
- 42. GDP, GSDP & AADT RELATIONSHIP42 Variable Equation for Karjan toll R² RMSE MAE MAPE GDPIND LN(AADT) = 0.22LN(GDPIND) + 9.04 0.72441 4695 3298 3.08% GSDPGUJ LN(AADT) = 0.2LN(GSDPGUJ) + 9.75 0.74658 4496 3203 3.00% GSDPMH LN(AADT) = 0.21LN(GSDPMH) + 9.55 0.71941 4728 3377 3.16% GDPIND - GSDPGUJ LN(AADT) = 2.79LN(GDPIND) - 2.32LN(GSDPGUJ) 0.99996 7667 6318 6.05% GSDPGUJ - GSDPMH LN(AADT) = -5.9LN(GSDPGUJ) + 6.68LN(GSDPMH) 0.99973 18279 16113 16.21% GDPIND - GSDPMH LN(AADT) = 4LN(GDPIND) - 3.6LN(GSDPMH) 0.99998 6163 5147 4.85% Variable Equation for Choryasi toll R² RMSE MAE MAPE GDPIND LN(AADT) = 0.22LN(GDPIND) + 8.87 0.82618 3243 2634 2.72% GSDPGUJ LN(AADT) = 0.21LN(GSDPGUJ) + 9.62 0.83755 3121 2562 2.66% GSDPMH LN(AADT) = 0.21LN(GSDPMH) + 9.4 0.82725 3219 2671 2.76% GDPIND - GSDPGUJ LN(AADT) = 2.77LN(GDPIND) - 2.29LN(GSDPGUJ) 0.99997 5729 5010 5.21% GSDPGUJ - GSDPMH LN(AADT) = -5.94LN(GSDPGUJ) + 6.71LN(GSDPMH) 0.99977 15469 13375 14.44% GDPIND - GSDPMH LN(AADT) = 3.92LN(GDPIND) - 3.52LN(GSDPMH) 0.99998 5549 4204 4.31%
- 43. CONTD.43 Variable Equation R² RMSE MAE MAPE GDPIND LN(AADT) = 0.35LN(GDPIND) + 7.29 0.86467 4100 3288 3.90% GSDPGUJ LN(AADT) = 0.31LN(GSDPGUJ) + 8.49 0.87359 4004 3086 3.64% GSDPMH LN(AADT) = 0.33LN(GSDPMH) + 8.14 0.84806 4338 3507 4.17% GDPIND - GSDPGUJ LN(AADT) = 2.38LN(GDPIND) - 1.81LN(GSDPGUJ) 0.99997 5395 4544 5.61% GSDPGUJ - GSDPMH LN(AADT) = - 4.89LN(GSDPGUJ) + 5.72LN(GSDPMH) 0.99978 12768 11082 14.08% GDPIND - GSDPMH LN(AADT) = 3.45LN(GDPIND) - 2.97LN(GSDPMH) 0.99999 2826 2324 2.70% Variable Equation R² RMSE MAE MAPE GDPIND LN(AADT) = 0.34LN(GDPIND) + 7.29 0.91217 3121 2561 3.10% GSDPGUJ LN(AADT) = 0.31LN(GSDPGUJ) + 8.48 0.92252 2938 2426 2.93% GSDPMH LN(AADT) = 0.33LN(GSDPMH) + 8.12 0.90280 3275 2673 3.24% GDPIND - GSDPGUJ LN(AADT) = 2.38LN(GDPIND) -1.81LN(GSDPGUJ) 0.99997 4828 4150 5.18% GSDPGUJ - GSDPMH LN(AADT) = - 4.95LN(GSDPGUJ) + 5.77LN(GSDPMH) 0.99980 11760 9973 13.15% GDPIND - GSDPMH LN(AADT) = 3.42LN(GDPIND) - 2.93LN(GSDPMH) 0.99999 2999 2554 3.10%
- 44. CONTD.44 80000 90000 100000 110000 80000 90000 100000 110000 PredictedAADT Observed AADT Predicted vs Observed Karjan AADT Ind Guj MH Ind-Guj Guj-MH Ind-MH 70000 80000 90000 100000 110000 70000 80000 90000 100000 110000 PredictedAADT Observed AADT Predicted vs Observed Choryasi AADT Ind Guj MH Ind-Guj Guj-MH Ind-MH 60000 70000 80000 90000 100000 110000 60000 80000 100000 PredictedAADT Observed AADT Predicted vs Observed Boriach AADT Ind Guj MH Ind-Guj Guj-MH Ind-MH 60000 70000 80000 90000 100000 110000 60000 70000 80000 90000 100000 110000PredictedAADT Observed AADT Predicted vs Observed Bhagwada AADT Ind Guj MH Ind-Guj Guj-MH Ind-MH
- 45. TIME SERIES45 Regression Model Moving Average with Classical Decomposition ARIMA Model
- 46. REGRESSION MODEL 46 Vehicle KARJAN Trend R² MAE MAPE RMSE Car Calibration y = 0.74x + 3691 0.55 1722 13 2236 Validation 2614 12 3522 LCV Calibration y = 0.85x + 596 0.76 203 6 251 Validation 331 8 411 Bus_Truck Calibration y = 0.7x + 1587 0.48 299 6 386 Validation 388 7 457 MAV Calibration y = 0.6x + 3603 0.38 386 4 501 Validation 821 8 896 Vehicle BORIACH Trend R² MAE MAPE RMSE Car Calibration y = 0.85x + 1196 0.72 1094 15 1448 Validation 1444 12 1961 LCV Calibration y = 0.97x + 157 0.92 215 6 276 Validation 304 6 384 Bus_Truck Calibration y = 0.79x + 931 0.60 247 6 334 Validation 242 6 346 MAV Calibration y = 0.78x + 1598 0.56 331 5 418 Validation 482 6 594
- 47. CONTD.47 Vehicle BHAGWADA Trend R² MAE MAPE RMSE Car Calibration y = 0.89x + 706 0.79 891 17 1250 Validation 1202 16 1581 LCV Calibration y = 0.95x + 235 0.90 276 7 361 Validation 230 4 1136 Bus_Truck Calibration y = 0.59x + 1845 0.34 268 6 339 Validation 243 5 1026 MAV Calibration y = 0.73x + 1915 0.51 378 6 469 Validation 245 3 77
- 48. MOVING AVERAGE WITH CLASSICAL DECOMPOSITION48 SEASONAL FACTOR CALCULATION 2009 2010 2011 2012 2013 2014 2015 Average Jan 1.03 1.06 1.06 1.08 0.97 1.06 1.04 1.04 Feb 0.99 1.01 1.03 1.01 0.97 1.27 1.05 1.04 Mar 0.91 0.86 0.92 0.94 0.88 1.05 0.93 0.93 Apr 0.99 0.96 1.08 1.01 0.88 1.08 1.00 1.00 May 1.38 1.30 1.22 1.34 1.26 1.26 1.29 1.29 Jun 1.11 1.06 1.09 1.09 1.18 1.00 1.09 1.09 Jul 0.85 0.86 0.83 0.79 0.89 0.76 0.83 0.83 Aug 0.75 0.83 0.88 0.85 0.94 0.82 0.85 0.84 Sep 0.80 0.72 0.72 0.75 0.73 0.78 0.75 0.75 Oct 1.22 0.81 1.01 0.76 0.79 1.00 0.93 0.93 Nov 1.04 1.37 1.15 1.29 1.34 0.97 1.19 1.19 Dec 1.15 1.14 1.05 1.09 1.05 1.04 1.08 1.08 Total 12.03 12 Required Total 12 Adjusting factor 0.998
- 49. CONTD.49 Vehicle Approach KARJAN Trend in data MAE RMSE MAPE Car Multiplicative Calibration y= 115.43x + 8733 885 1333 7 Validation 1321 2376 6 Additive Calibration y =114.03x + 8806 752 1014 6 Validation 1020 1570 5 LCV Multiplicative Calibration y= 18.52x + 2888 228 316 7 Validation 412 490 9 Additive Calibration y =18.23x + 2867 154 200 4 Validation 258 352 6 BUS_TRUCK Multiplicative Calibration y= -10.73x + 5634 361 452 7 Validation 1324 1379 22 Additive Calibration y= -10.55x + 5607 229 291 4 Validation 1340 1350 23 MAV Multiplicative Calibration y=15.95x + 8303 555 727 6 Validation 731 906 7 Additive Calibration y= 15.56x + 8264 347 422 4 Validation 823 884 8
- 50. CONTD.50 5000 10000 15000 20000 25000 5000 10000 15000 20000 25000 Predicted Observed Cars M_Predicted A_Predicted 2000 3000 4000 5000 2000 3000 4000 5000 Predicted Observed LCV M_Predicted A_Predicted 3000 4000 5000 6000 7000 3000 4000 5000 6000 7000 Predicted Observed BUS_TRUCK M_Predicted A_Predicted 6000 7000 8000 9000 10000 11000 12000 6000 7000 8000 9000 10000 11000 12000 Predicted Observed MAV M_Predicted A_Predicted
- 51. CONTD.51 Vehicle Approach BORIACH Trend in data MAE RMSE MAPE Car Multiplicative Calibration y = 98.69x + 3263 792 1010 11 Validation 1810 2170 16 Additive Calibration y = 98.45x + 3253 822 1053 12 Validation 1879 2082 17 LCV Multiplicative Calibration y = 38.66x + 1842 95 242 6 Validation 646 767 13 Additive Calibration y = 38.70x + 1835 214 245 6 Validation 648 754 13 BUS_TRUCK Multiplicative Calibration y = -1.26x + 4317 440 472 10 Validation 352 466 8 Additive Calibration y = -1.17x + 4311 407 471 10 Validation 353 466 8 MAV Multiplicative Calibration y = 20.08x + 6188 524 248 3 Validation 563 641 6 Additive Calibration y = 18.45x + 6236 199 254 3 Validation 654 721 7
- 52. CONTD.52 0 5000 10000 15000 0 5000 10000 15000 Predicted Observed Cars M_Predicted A_Predicted 1000 2000 3000 4000 5000 6000 7000 1000 2000 3000 4000 5000 6000 7000 Predicted Observed LCV M_Predicted A_Predicted 3000 4000 5000 6000 3000 4000 5000 6000 Predicted Observed BUS_TRUCK M_Predicted A_Predicted 5000 6000 7000 8000 9000 10000 11000 5000 6000 7000 8000 9000 10000 11000 Predicted Observed MAV M_Predicted A_Predicted
- 53. CONTD.53 Vehicle Approach BHAGWADA Trend in data MAE RMSE MAPE Car Multiplicative Calibration y = 89.96x + 1635 1010 1458 16 Validation 3165 3404 43 Additive Calibration y = 90.44x + 1576 1065 1487 19 Validation 3170 3356 45 LCV Multiplicative Calibration y =45.65x + 1825 264 318 7 Validation 960 1091 18 Additive Calibration y =45.7x + 1822 264 315 7 Validation 969 1090 18 BUS_TRUCK Multiplicative Calibration y= -4.99x + 4851 338 912 6 Validation 542 597 11 Additive Calibration y = -4.24x + 4819 341 911 6 Validation 519 578 11 MAV Multiplicative Calibration y=19.97x + 5903 219 266 3 Validation 357 464 4 Additive Calibration y =19.95x + 5908 218 265 3 Validation 353 453 4
- 54. CONTD.54 0 5000 10000 15000 0 5000 10000 15000 Predicted Observed Cars M_Predicted A_Predicted 1500 3000 4500 6000 1500 3000 4500 6000 Predicted Observed LCV M_Predicted A_Predicted 3000 4000 5000 6000 7000 3000 4000 5000 6000 7000 Predicted Observed BUS_TRUCK M_Predicted A_Predicted 5000 6000 7000 8000 9000 5000 6000 7000 8000 9000 Predicted Observed MAV M_Predicted A_Predicted
- 55. ARIMA MODEL55
- 56. CONTD.56 ARIMA(p,d,q) ‘p’ represent Autoregression (AR), ‘d’ represent non-seasonal differencing order, and ‘q’ represent Moving average (MA) Seasonal ARIMA(p,d,q)x(P,D,Q) (SARIMA) ‘P’ represent Seasonal Autoregression (SAR) order, ‘D’ represent Seasonal differencing order and ‘Q’ represent Seasonal Moving average (SMA) ACF PACF AR Geometric decay Significant till p lags MA Significant till p lags Geometric decay
- 57. AR SIGNATURE57
- 58. MA SIGNATURE58
- 59. SAR AND SMA SIGNATURE59 SAR signature: Positive spikes in ACF at lag s, 2s, 3s and positive spike in PACF at lag s. SMA signature: Negative spike in ACF at lag s, and negative spikes in PACF at lags s, 2s, 3s.
- 60. PROCEDURE60 Analysis involves formation of ACF and PACF plots for four different cases which are: Plots without any differencing. Plots with non-seasonal differencing. Plots with seasonal differencing. Plots with non-seasonal and seasonal differencing.
- 61. CONTD.61 For example: Case with Non-Seasonal and Seasonal Differencing of Karjan toll Car traffic Model – ARIMA(p,d,q)x(P,D,Q) = ARIMA(0,1,1)x(0,1,1)
- 62. SPSS 2362
- 63. KARJAN TOLL 63 Vehicle Model R2 RMSE MAPE MAE Normalized BIC CAR ARIMA (011 010) Calibration .722 1598 8 1131 14.81 Validation 2386 8 1782 ARIMA (011 011) Calibration .823 1283 7 949 14.44 Validation 2127 8 1728 LCV ARIMA (011 011) Calibration .717 223 5 176 10.94 Validation 97 2 97 ARIMA (011 000) Calibration .767 248 6 199 11.08 Validation 504 10 454 BUS_TRUCK ARIMA (100 010) Calibration .507 329 5 238 11.65 Validation 799 13 761 ARIMA (011 011) Calibration .627 289 4 208 11.46 Validation 718 12 699 MAV ARIMA (011 000) Calibration .360 514 5 419 12.54 Validation 1277 10 1131 ARIMA (011 011) Calibration .328 502 4 384 12.56 Validation 1123 10 1049
- 64. CONTD.64
- 65. BORIACH TOLL 65 Vehicle Model R2 RMSE MAPE MAE Normalized BIC CAR ARIMA (011 011) Calibration .867 1022 9 701 13.98 Validation 1600 13 1409 ARIMA (200 011) Calibration .865 1039 9 706 14.07 Validation 1279 8 950 LCV ARIMA (011 000) Calibration .936 245 5 189 11.06 Validation 427 6 319 ARIMA (011 010) Calibration .902 285 5 195 11.37 Validation 567 9 463 BUS_TRUCK ARIMA (100 010) Calibration .289 448 6 263 12.27 Validation 734 12 536 ARIMA (011 000) Calibration .577 345 6 247 11.74 Validation 629 16 656 MAV ARIMA (011 010) Calibration .494 400 4 298 12.04 Validation 584 5 435 ARIMA (011 000) Calibration .608 385 4 293 11.96 Validation 574 5 470
- 66. CONTD.66
- 67. BHAGWADA TOLL 67 Vehicle Model R2 RMSE MAPE MAE Normalized BIC CAR ARIMA (010 011) Calibration .830 1126 11 693 14.11 Validation 1268 12 958 ARIMA (100 011) Calibration .833 1124 11 701 14.17 Validation 1766 14 1142 LCV ARIMA (011 000) Calibration .917 334 6 246 11.67 Validation 516 8 408 ARIMA (200 011) Calibration .891 346 6 238 11.87 Validation 968 17 866 BUS_TRUCK ARIMA (011 011) Calibration .682 246 4 193 11.13 Validation 530 9 431 ARIMA (200 011) Calibration .694 244 4 187 11.18 Validation 469 8 370 MAV ARIMA (011 000) Calibration .619 410 5 328 12.08 Validation 523 6 432 ARIMA (011 010) Calibration .472 462 5 353 12.33 Validation 1127 13 1009
- 68. CONTD.68
- 69. VOLUME TO CAPACITY ANALYSIS69 LOS THRESHOLD FOR 6-LANE DIVIDED INTERURBAN HIGHWAY SEGMENTS LOS V/C PCU/day Threshold A <0.2 <27000 34000 @ LOS-B: Suggested threshold flow for conversion from six lane to eight lane divided road to ensure enhanced safety in traffic operations. B 0.21-0.3 27001-41000 C 0.31-0.5 41001-68000 D 0.51-0.7 68001-95000 E 0.7-1 95001-136000 F >1 >136000 Indian Highway Capacity Manual (Indo-HCM)
- 70. KARJAN & CHORYASI TOLL70 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Volume to Capacity Analysis REmp RIn RegTV RegV 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Volume to Capacity Analysis REmp RIn RegTV RegV 0.0 0.5 1.0 1.5 2.0 Volume to Capacity Analysis GDPIND GSDPGUJ GSDPMH GDPIND - GSDPGUJ GSDPGUJ - GSDPMH GDPIND - GSDPMH 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Volume to Capacity Analysis GDPIND GSDPGUJ GSDPMH GDPIND - GSDPGUJ GSDPGUJ - GSDPMH GDPIND - GSDPMH
- 71. BORIACH & BHAGWADA TOLL71 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Volume to Capacity Analysis REmp RIn RegTV RegV 0.0 0.5 1.0 1.5 2.0 Volume to Capacity Analysis GDPIND GSDPGUJ GSDPMH GDPIND - GSDPGUJ GSDPGUJ - GSDPMH GDPIND - GSDPMH 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Volume to Capacity Analysis REmp RIn RegTV RegV 0.0 0.5 1.0 1.5 2.0 Volume to Capacity Analysis GDPIND GSDPGUJ GSDPMH GDPIND - GSDPGUJ GSDPGUJ - GSDPMH GDPIND - GSDPMH
- 72. TIME SERIES MODELS72 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 2019-20 2020-21 2021-22 2022-23 2023-24 2024-25 2025-26 2026-27 2027-28 2028-29 2029-30 2030-31 2031-32 2032-33 2033-34 V/C Year Volume to Capacity Analysis Karjan Boriach Bhagwada
- 73. 73 DIVERSION
- 74. Diversion Analysis based on Origin-Destination Primary Survey Data Survey Proportion Percentage Actual Volume Proportion Average Volume Calculation of Utility value Diversion Proportion (Logit Model) Highway Traffic MADT for NH48 Toll Plazas MADT for VME sections DIVERSION ANALYSIS74
- 75. Karjan Toll Narmada Toll Choryasi Toll Boriach Toll Bhagwada Toll National Highway 48 External East External South East External North External South Vadodara External West Bharuch Surat Navsari Valsad 1 2 3 4 5 6 EXISTING HIGHWAY (NH-48)75
- 76. MONTH KARJAN TOLL NARMADA TOLL CHORYASI TOLL BORIACH TOLL BHAGWADA TOLL Sep-18 1,17,596 1,21,835 1,10,531 1,18,284 1,03,961 Oct-18 1,20,453 1,24,776 1,14,687 1,21,540 1,04,342 Nov-18 1,24,255 1,21,260 1,09,491 1,15,035 98,709 Dec-18 1,27,274 1,29,522 1,22,924 1,26,250 1,07,934 Jan-19 1,27,993 1,27,935 1,31,608 EXTERNAL EAST 1,33,347 EXTERNAL SOUTHEAST 1,09,985 Feb-19 1,31,942 1,33,302 1,47,885 1,40,237 1,20,986 EXTERNAL NORTH EXTERNAL SOUTH VADODARA EXTERNAL WEST BHARUCH SURAT NAVSARI VALSAD 76
- 77. Sep-18 Oct-18 Nov-18 Dec-18 Jan-19 Feb-19 80,000 90,000 1,00,000 1,10,000 1,20,000 1,30,000 1,40,000 1,50,000 Traffic Volume (AADT) variation over Study Stretch Sep-18 Oct-18 Nov-18 Dec-18 Jan-19 Feb-19 TOLLS TRAFFIC VARIATION OVER STUDY STRETCH 77
- 78. V/C FOR TRAFFIC AT TOLL PLAZAS OF NH48 UPTO HORIZON YEAR 2035 78 0.0 0.5 1.0 1.5 2.0 2018 2023 2028 2033 V/C Year V/C Trend for NH-48 Karjan Narmada Choryasi Boriach Bhagwada LOS B LOS C LOS F
- 79. National Highway 48 External East External SouthEast External North External South Vadodara External West Bharuch Surat Navsari Valsad Vadodara Mumbai Expressway External East External SouthEast External North External South Vadodara External West Bharuch Surat Navsari Valsad 1 2 3 4 5 6 1 2 3 4 5 6 7 EXISTING ROUTE (NH-48) WITH ALTERNATE FACILITY (VME)79
- 80. 80 Binary Logit model based on vehicle category. P ΤR2 R1 k = eUR2 eUR1 + eUR2 Utility Functions derived from Logit Model. Uk R1 = α H_TTR1+ β H_costR1 + γ H_distR1 + C Uk R2 = α H_TTR2+ β H_costR2 + γ H_distR2 + C ROUTE CHOICE MODEL Where, Pk (R2/R1) = Probability of shifting from route 1 to 2 for kth vehicle category. Uk R1 = Utility function of route 1 for kth vehicle category. Uk R2 = Utility function of route 2 for kth vehicle category. H_costR1/ H_costR1 = Travel cost for route 1 and route 2. H_TTR1 / H_TTR2 = Travel time for route 1 and route 2. H_distR1 / H_distR2 = Distance for route 1 and route 2. C = Constant for unexplained part. α, β, γ = Parameters or coefficients of the variables in model.
- 81. DATA FORMATION81
- 82. 82
- 83. BINARY LOGIT MODEL83 Vehicle Utility Equation -2 Log Likelihood Nagelkerke R² Percentage Correct Car Ucar = -0.016 H_TT+ 0.042 H_cost – 0.083 H_dist 316.153 .675 85.6 LCV ULCV = -0.012 H_TT + 0.042 H_cost – 0.182 H_dist 86.434 .803 90.3 Bus_Truck UBus_Truck = -0.013 H_TT + 0.011 H_cost – 0.057H_dist 367.997 .524 76.4 MAV UMAV= -0.019 H_TT + 0.009 H_cost – 0.072H_dist 682.999 .555 80.5
- 84. UTILITY EQUATION DATA84 Utility Modelling • Using primary survey data for each vehicle category. • Each data set was converted into 1-4 choice sets based on responses of respondent for Willingness to pay section. Logit Model inputs for Utility Calculation Distance Based on Entry and Exit point of highway for each O-D pair NH48 Distance between Entry & Exit point of NH48 alignment VME Distance between Entry & Exit point of VME proposed alignment Travel Time for each O-D pair NH48 Travel time on existing route through primary survey data (50 percentile value) VME Travel time through 10% increase in journey speed (Assumption) Travel Cost for each O-D pair NH48 Travel cost on existing route through toll collection VME Toll Cost through document with Regd. No. D.L. - 33004/99, The Gazette of India: Extraordinary, Part II Section - 3, MORTH Vehicle Car LCV Bus_Truck MAV NH Speed (KM/hr) 73 64 59 55 VME Speed (KM/hr) 80 70 65 60 Toll Rate per KM 1.08 1.92 3.92 6.24 Toll Rate per Structure 5 7.5 15 22
- 85. DIVERSION PROPORTION85 O-D Pair Distance Vehicle Category Car LCV Bus_Truck MAV External North External South 260 0.92 0.99 0.89 0.93 External North External South East 250 0.92 0.97 0.93 0.96 External North External East 161 0.76 0.85 0.73 0.77 External North Bharuch 91 0.61 0.58 0.63 0.65 External North Surat 124 0.6 0.71 0.5 0.49 External North Navsari 188 0.85 0.92 0.85 0.9 External North Valsad 250 0.11 0.01 0.21 0.13 Vadodara Vadodara 0 0 0 0 0 Vadodara Bharuch 91 0.61 0.58 0.63 0.65 Vadodara Surat 124 0.6 0.71 0.5 0.49 Vadodara Navsari 188 0.23 0.07 0.29 0.22 Vadodara Valsad 250 0.11 0.01 0.21 0.13 External West External South 169 0.87 0.98 0.82 0.87 External West External South East 159 0.88 0.96 0.88 0.93 External West External East 70 0.66 0.8 0.61 0.65
- 86. CONTD.86 O-D Pair Distance Vehicle Category Car LCV Bus_Truck MAV External West Surat 33 0.48 0.63 0.37 0.34 External West Navsari 97 0.16 0.05 0.19 0.13 External West Valsad 159 0.07 0.01 0.14 0.07 Bharuch Bharuch 0 0 0 0 0 Bharuch Surat 33 0.48 0.63 0.37 0.34 Bharuch Navsari 97 0.16 0.05 0.19 0.13 Bharuch Valsad 159 0.07 0.01 0.14 0.07 Surat Surat 0 0 0 0 0 Surat Navsari 64 0.17 0.03 0.29 0.22 Surat Valsad 126 0.08 0 0.21 0.13 External East External South 99 0.77 0.92 0.75 0.78 External East External South East 89 0.79 0.86 0.82 0.87 External East Vadodara 161 0.76 0.85 0.73 0.77 External East Bharuch 70 0.66 0.8 0.61 0.65 External East Navsari 27 0.09 0.01 0.13 0.07
- 87. CONTD.87 O-D Pair Distance Vehicle Category Car LCV Bus_Truck MAV External East Valsad 89 0.04 0 0.09 0.04 Navsari Navsari 0 0 0 0 0 Navsari Valsad 62 0.02 0 0.04 0.02 External South East External South 10 0.47 0.66 0.39 0.35 External South East Vadodara 250 0.92 0.97 0.93 0.96 External South East Bharuch 159 0.88 0.96 0.88 0.93 External South East Surat 126 0.21 0.03 0.33 0.27 External South East Navsari 62 0.67 0.76 0.68 0.72 External South East Valsad 0 0 0 0 0 Valsad Valsad 0 0 0 0 0 External South Vadodara 260 0.92 0.99 0.89 0.93 External South Bharuch 169 0.87 0.98 0.82 0.87 External South Surat 136 0.2 0.06 0.24 0.16 External South Navsari 72 0.66 0.85 0.58 0.58 External South Valsad 10 0.74 0.95 0.6 0.62
- 88. LOS THRESHOLDS88 LOS Thresholds for Six Lane Divided Interurban Expressway Segments LOS V/C PCU/day Threshold A <0.25 <39800 58200 @ LOS-B: Suggested threshold flow for conversion from six lane to eight lane divided road to ensure enhanced safety in traffic operations. B 0.26-0.5 39801- 76500 C 0.51-0.75 76501- 114800 D 0.76-0.93 114801- 142300 E 0.94-1 142301- 153000 F >1 >153000 LOS Thresholds for Eight Lane Divided Urban Expressway Segments LOS V/C PCU/day Threshold A <0.25 <47600 69600 @ LOS-B: Suggested threshold flow for conversion from six lane to eight lane divided road to ensure enhanced safety in traffic operations. B 0.26-0.5 47601- 91500 C 0.51- 0.75 91501- 137300 D 0.76- 0.93 137301- 170200 E 0.94-1 170201- 183000 F >1 >183000
- 89. VOLUME TO CAPACITY ANALYSIS89 0.0 0.5 1.0 1.5 2.0 2018 2023 2028 2033 V/C Year V/C Trend for NH-48 Karjan Narmada Choryasi Boriach Bhagwada LOS B LOS C LOS F
- 90. CONTD.90 0.0 0.5 1.0 1.5 2.0 2018 2022 2026 2030 2034 V/C Year V/C Trend for 6-Lane VME 1 2 3 4 5 6 7 LOS B LOS C LOS F 0.0 0.5 1.0 1.5 2.0 2018 2022 2026 2030 2034 V/C Year V/C Trend for 8-Lane VME 1 2 3 4 5 6 7 LOS B LOS C LOS F
- 91. RESULTS AND DISCUSSIONS The employment multiplier for region is 1.86. i.e. E = 1.86 EB.. Employment multiplier of our districts in study regions i.e. Vadodara, Bharuch, Surat, Navsari and Valsad are 1.54, 1.53, 2.42, 1.73 and 1.75 respectively with 1.8 value for Gujarat State. The actual growth rate of investment and employment in region are 29% and 21% respectively. The traffic of NH-48 is forecasted using aggregate and disaggregate models. Aggregate models are developed with annual economic data that are Employment and investment data of region, Employments models are performing better 91
- 92. CONTD. Vehicle registration data of Gujarat, and Vehlice registration data is able to explain traffic well GDPIND, GSDPGUJ and GSDPMH data. GSDPGUJ model is performing well Disaggregate models are developed using Regression Analysis Bus_Truck and MAV models are not able to explain data Moving Average with Classical Decomposition Bus_Truck and Car models have high MAE and MAP values ARIMA Models Most of the models have seasonal Moving average component after seasonal decomposition. 92
- 93. CONTD. V/C Analysis RegV, RegTV and GSDPGUJ models are showing smooth growth trend while others are shooting up after 2019. The highway sections have already reached capacity and the traffic is except to grow twice of capacity by 2035. Diversion Analysis The diversion proportion through logit model show that mostly E-E (through) traffic, I-E and E-I traffic will shift to proposed facility (VME). The V/C analysis show that NH-48 and VME will operate in LOS ‘C’, considering VME as 6-lane divided while VME can also be in LOS ‘B’ for 8-lane divided. 93
- 94. FUTURE SCOPE OF WORK Change in land use pattern which is responsible for development traffic for NH during the horizon year is likely to be considered. Freight demand modeling is to be included to calculate the growth rate of freight vehicles for exclusive freight corridors. 94
- 95. REFERENCES Blien, U. & Tassinopoulos, A., 2001. Forecasting Regional Employment with the ENTROP Method. Regional Studies, 35(2), pp. 113-124. Dan, W. & Yangshan, T., 2012. The Application of System Dynamic Model for Freeway Traffic Volume Forecasting.. s.l., s.n. Fang, Z., 2007. Input-output model for forecasting the interregional freight volume.. pp. 2846-2851. IRC:108-2015, 2015. Guidelines for Traffic Forecast on Highways. New Delhi: Indian Road Congress. Jha, K., Ponnu, B. & Arkatkar, S., 2012. Time Series Analysis: A Contemporary Approach to Traffic Volume Forecasting.. pp. 221-225. Jin, Y. & Williams, I. N., 2000. A New Regional Economic Model for European Transport Corridor Studies.. Marcial Echenique and Partners, Cambridge, pp. 149-169. Kim, J. Y. & Han, J. H., 2015. Straw effects of new highway construction on local population and employment growth. Habitat International, Volume 53, pp. 123-132. Lewis, C. D., 1982. Industrial and business forecasting methods: A practical guide to exponential smoothing and curve fitting. Luo, L., 2008. Research on the Land Use Model in the Regional Traffic Demand Forecast.. pp. 144-150. Pradhan, R. P. and Bagchi, T. P. (2013) ‘Effect of transportation infrastructure on economic growth in India: The VECM approach’, Research in Transportation Economics. Elsevier Ltd, 38(1), pp. 139–148. doi: 10.1016/j.retrec.2012.05.008. 95
- 96. CONTD. Raghuram, G. and Ravi Babu, M. (1999) ‘Alternate means of financing railways’, Vikalpa, 24(1), pp. 13–25. doi: 10.1177/0256090919990103. Sanchez-Roble, B. (1994) ‘Infrastructure Investment and Growth : Some Empirical Evidence’, Contemporary Economic Policy. Short, J. and Kopp, A. (2005) ‘Transport infrastructure: Investment and planning. Policy and research aspects’, Transport Policy, 12(4), pp. 360–367. doi: 10.1016/j.tranpol.2005.04.003. Sorratini, J. A., 2000. Estimating statewide truck trips using commodity flows and input-output coefficient.. Journal of Transportation & Statistics, pp. 53-67. Stein, M. M., 1975. Regional Impacts of national transport systems on population and travel. Journal of Transport Economics and Policy, pp. 255-267. Thabassum, S., 2013. Impact of state-wise vehicle contribution on traffic growth rates for National Highways. International Journal for Engineering Research and technology, 2(10). Wang, B., 2015. Estimating Economics impact of transport investment using TREDIS: A case study on a National Highway upgrade program.. s.l., s.n. Wardrop, J. G., 1952. Some Theoretical Aspects of Road Traffic Research. s.l., s.n., pp. 325-362. Williams, B. M. & Hoel, L. A., 2003. Modeling and Forecasting Vehicular Traffic Flow as a Seasonal ARIMA Process: Theoretical Basis and Empirical Results. Journal of Transportation Engineering, 129(6), pp. 664- 672. 96
- 97. CONTD.97 Armstrong, J. S. and Collopy, F. (1992) ‘Error measures for generalizing about forecasting methods: Empirical comparisons’, International Journal of Forecasting, 8(1), pp. 69–80. doi: 10.1016/0169- 2070(92)90008-W. Carbone, R. and Armstrong, J. S. (1982) ‘Evaluation of Extrapolative Forecasting Methods : Results of a Survey of Academicians and Practitioners’, I(October 1981), pp. 215–217. Esfahani, H. S. and Ramírez, M. T. (2003) ‘Institutions, infrastructure, and economic growth’, Journal of Development Economics, 70(2), pp. 443–477. doi: 10.1016/S0304-3878(02)00105-0. Fretz, S. and Gorgas, C. (2013) ‘Regional economic effects of transport infrastructure expansions : Evidence from the Swiss highway network’, (April), pp. 1–28. Fricker, J. D. and Saha, S. K. (1988) ‘Traffic Volume Forecasting Methods for Rural State Highways: Publication FHWA/IN/JHRP-86/20’, Transportation reseach Record, 1203, pp. 10–26. Mirwaldt, K., Mcmaster, I. and Bachtler, J. (2005) Reconsidering Cohesion Policy : The Contested Debate on Territorial Cohesion, European Policies Research Centre. Naylor, T. H., Seaks, T. G. and Wichern, D. W. (1972) ‘Box-Jenkins Methods: An Alternative to Econometric Models’, International Statistical Review / Revue Internationale de Statistique, 40(2), p. 123. doi: 10.2307/1402755. Phang, S. Y. (2002) ‘Strategic Infrastructure Investment Decisions in Airport and Rail Development: The Case of Singapore’, Singapore Management University, 10, pp. 27–33. Available at: http://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=2899&context=soe_research.
- 99. THANK YOU