CHAPTER 2

Population Forecasting for Water Supply Design

Population Forecasting

Specifies methods for forecasting design population over the 30-year design horizon of water supply works — arithmetical, geometric, incremental, graphical, logistic, and comparative methods. Design population drives every subsequent design parameter.

📋 Planning & Design PeriodManual on Water Supply and Treatment3rd Edition (1999) with 2024 revision updates

Key formulas

Arithmetical increase method: P_n = P_0 + n × X, where X = average decadal increase, n = decades from base year.
Geometric increase method: P_n = P_0 × (1 + r/100)^n, where r = average growth rate per decade, n = decades.
Incremental increase method: P_n = P_0 + n × X + [n(n+1)/2] × Y, where X = avg increase, Y = avg incremental increase of increase.
Logistic curve (for saturating populations): P_t = P_s / (1 + m × e^(-k × t)), where P_s = saturation population, m and k = constants from past data.

Key values & thresholds

design horizon years: 30
reassessment interval years: 10
census years required: Last 3-5 decadal census readings
typical urban growth rate pct decade: 20-35
typical metro growth rate pct decade: 15-25 (maturing)
typical new town growth rate pct decade: 40-80

Clause-level requirements

Forecast methodology is selected based on city maturity: rapidly growing (geometric), steady growth (incremental/arithmetical), mature/saturating (logistic).
Minimum 3 past census decadal populations required; 5 or more preferred for reliability.
Floating population (workers, students, tourists, daily commuters) must be added — can be 10-30% of permanent population in Indian cities.
Industrial/commercial population equivalent: convert industrial water demand to equivalent population using 30 lpcd per worker for factories, 45 lpcd for offices.
Saturation population (for logistic method) = f(land area × assumed density) — e.g., 400 persons/hectare for medium-density Indian urban development.

Practitioner notes — what goes wrong in the field

Geometric method typically gives highest forecast — use for new/rapidly growing cities (Gurgaon, Navi Mumbai, Bangalore IT corridors, Noida Extension).
Incremental method suitable for most Indian tier-2 cities with steady growth trajectory.
Logistic method underutilized but correct choice for saturating metros — Mumbai Island City, Kolkata, Chennai are approaching saturation at ~500 persons/hectare.
Always check census data against NPR (National Population Register) and municipal property records — census undercount common in slum/informal areas.
Floating population in major cities: Delhi +25%, Mumbai +20%, Bangalore +18% over permanent. Include in water demand estimation.
Tourist-heavy destinations: Goa, Agra, Jaipur, Varanasi — peak tourist season adds 50-200% to base population; design peak works for tourist peak.
Industrial clusters: separate demand projection for industrial estates (ONGC, JNPT, Jamnagar, Sriperumbudur) — industrial lpcd can exceed 500 per worker for water-intensive plants.
Most common error: using only one method and not cross-checking. Best practice: run all 4 methods, compare results, use highest rational projection (usually geometric for growing cities).

FAQs

Which forecasting method should I use?

Depends on city maturity: arithmetical for slow-growing old cities; geometric for rapidly growing/new cities; incremental for steady growth; logistic for saturating metros. Run all methods, use highest rational projection.

What if past census data is missing?

Use Planning Commission / NITI Aayog urban population estimates, state urban development plans, or municipal voter rolls. Minimum 3 data points required — fewer makes projection unreliable.

Should floating population be included?

Yes — 10-30% addition typical for Indian cities. Floating population drinks water, uses toilets, generates wastewater exactly like permanent residents. Ignoring this underestimates demand by 15-25%.

What is typical urban decadal growth rate?

Mature metros 15-25% (Mumbai, Chennai, Kolkata); growing tier-1 25-40% (Delhi, Bangalore, Hyderabad); new towns and IT corridors 40-80% (Gurgaon, Navi Mumbai, Hitec City).

Calculator

Prefer a dedicated page with worked example, FAQs, and cross-references? Open as standalone tool →

Population Forecasting (30-year design horizon)

↓ Download Excel

Project the base-year population to the design year using all 4 CPHEEO methods. Use the highest rational forecast (typically geometric for growing cities, logistic for saturating metros).

Inputs

Base-year populationpersons

Average decadal increasepersons

From past census data

Average increment of increasepersons

For incremental method; set 0 if unknown

Decadal growth rate%

Decades to forecastdecades

3 decades = 30-year design period

Saturation population (logistic)persons

City's ultimate saturation based on land area × density

Outputs

Arithmetical forecast

86,000persons

P = P₀ + n × X

Geometric forecast

97,656persons

P = P₀ × (1 + r/100)^n

Incremental forecast

98,000persons

P = P₀ + n × X + n(n+1)/2 × Y

Logistic forecast

1,42,190persons

P = Ps / (1 + m × e^(-k × n)) (m, k from past data)

↪ Simplified: uses m = 1, k = 0.3 typical

Design population (highest rational)

1,42,190persons

max of the 4 methods

CPHEEO Reference Values

Design horizon30 years (3 decades)

Reassessment interval10 years

Method selectionGrowing cities → geometric · Saturating metros → logistic

Download the Excel version to keep a local copy with live formulas — change inputs in the sheet and outputs recompute automatically.

Cross-references

CPHEEO WS Chapter 1Census of IndiaState Urban Development Plans