IRC 108:1996 is the Indian Standard (IRC) for guidelines for traffic prediction on rural highways. IRC 108:1996 provides methodology for traffic forecasting on rural highways — essential input for highway DPR, pavement design, structural design, land acquisition, and capacity planning. Forecasting horizon: 20 years (pavement), 30 years (structures), 50 years (land acquisition). Four methods: past-trend extrapolation (regression on historical counts), elasticity method (traffic growth = elasticity × GDP growth), gravity model (O-D analysis), and four-step model (comprehensive network analysis). Typical Indian growth rates: NH 5-8% annual, SH 3-5%, MDR 2-4%; commercial traffic 6-10%. Four-step model components: trip generation (land use-based), distribution (between zones), modal split (car/bus/rail), and assignment (to routes). Amendment No. 1 (2015) added induced traffic (10-25% induction for new highways in first 3 years) and climate-change impact considerations. Amendment No. 2 (2022) aligned with Bharatmala forecasting methodology (GDP-linked growth, multi-scenario). Accurate traffic forecasting is critical — 20-30% error (not uncommon) translates to wrong capacity/design decisions and project failure. IRC 108 methodology ensures systematic, defensible forecasts.
Specifies methodology for traffic forecasting on rural highways — including growth rate estimation, trip generation, trip distribution, modal split, and assignment for planning of new highways and upgrades.
- Status
- Current
- Usage level
- Essential
- Domain
- Transportation — Traffic Engineering / Planning
- Type
- Guidelines
- Amendments
- Amendment No. 1 (2015) — induced traffic (10-25% in first 3 years), climate-change impact; Amendment No. 2 (2022) — Bharatmala forecasting methodology alignment, multi-scenario GDP-linked
Also on InfraLens for IRC 108
Practical Notes
! Traffic forecasting errors: 20-30% common at 20-year horizon. Optimistic projections (common in PPP BOT) led to many failed concessionaires 2008-2015.
! Past-trend extrapolation: simple and defensible for stable growth. Breaks down during economic disruption (demonetization 2016, Covid 2020). Modern practice: combine with elasticity method.
! Elasticity method: GDP growth × elasticity coefficient. India GDP 6-8%, passenger elasticity 1.0 → 6-8% passenger traffic growth. Consistent with Indian observed growth.
! Freight elasticity > passenger (1.0-1.5 vs 0.8-1.2): as GDP rises, freight increases faster due to industrialization and trade. Relevant for truck-heavy corridors.
! Four-step model: most comprehensive. Used for: urban area transport studies, regional network planning, Bharatmala/Sagarmala corridor studies. Software: VISSIM, TransCAD, CUBE.
! Software tools: TransCAD, Visum, EMME/2 for comprehensive models. Excel-based for simple past-trend projections. Cost ₹5-50 lakh per license for commercial software.
! Induced traffic (Amendment No. 1, 2015): new highway generates additional trips beyond pure growth. 10-25% induction in first 3 years. Often overlooked in DPR — causes post-commissioning capacity issues.
! Vehicle class projections: 2W has slowest growth (saturation); cars growing fastest (middle-class expansion); trucks growing 6-10% (economic activity). Aggregate PCU different from vehicle count growth.
! Regional variation: Delhi-Mumbai corridor > 8% growth historical; Bihar-UP rural corridors < 4%; hill regions < 3%. Use region-specific rather than national averages.
! Tourist corridor forecasting: Himachal, Sikkim, Kerala seasonal peaks 3-5× monthly average. Include in capacity design, not just average.
! Pilgrim traffic: Haridwar-Rishikesh festival peaks 10-20× normal; Kedarnath June-October 100× winter. Special provisions, not routine design.
! Commercial vs agricultural corridors: Grand Trunk Road corridors growing 7-10% (industrial activity); pure agricultural corridors 3-5%. Economic base matters.
! Peak vs average: peak month / average month = 1.3-1.8 typical. Peak hour / average day = 8-12%. Both matter for capacity design.
! Covid impact (2020-21): traffic dropped 30-60% during lockdown; recovered within 12-18 months. Post-Covid projections use 2022+ data as baseline.
! Demonetization impact (2016-17): 20-30% drop for 6-9 months; full recovery by 2018. Long-run patterns unchanged.
! Economic growth sensitivity: 1% GDP change → 0.8-1.5% traffic change (elasticity). Multi-scenario forecasts essential for robust planning.
! For PPP BOT projects: conservative (low-side) projections to avoid concessionaire failure. Independent traffic consultant recommended. Sensitivity analysis on ±25% essential.
! Data quality: traffic counts per IRC 9 methodology essential. Manual counts ±10-15% error; electronic counters ±3-5% error. Quality data enables reliable forecasts.
! Post-commissioning verification: 2-5 year monitoring verifies design assumptions. Identifies need for capacity enhancements early.
! Cost of traffic forecasting study: ₹10-50 lakh for state highway; ₹1-5 crore for major NH / ring road studies. 0.5-2% of total project cost but critical for investment decision.