Part 1 of 2
Water Demand & Population Forecasting
Last reviewed 16 Sept 2026 · 7 min read
Planning a water supply scheme
A public water supply scheme must supply enough water of acceptable quality at adequate pressure to the population at the end of the design period. The first steps are to estimate the per capita demand, the future population and the variations in demand.
Types of water demand
| Demand | Includes |
|---|---|
| Domestic | Drinking, cooking, bathing, washing clothes and utensils, flushing toilets, gardening — usually the largest share in towns |
| Industrial | Factories, processing units (varies greatly with type of industry) |
| Institutional and commercial | Schools, hospitals, hotels, offices, restaurants, railway and bus stations |
| Public (civic) use | Street washing, sewer flushing, public parks, fountains |
| Fire demand | Fire fighting (small annual quantity but high rate) |
| Losses and wastage | Leakage from mains and fittings, unauthorised connections, meter errors — part of non-revenue water |
Per capita demand norms (CPHEEO)
| Classification of town/city | Recommended maximum water supply (litres per capita per day) |
|---|---|
| Towns with piped water supply but without sewerage | 70 |
| Cities with piped water supply where sewerage exists or is planned | 135 |
| Metropolitan and mega cities with piped supply and existing/planned sewerage | 150 |
| Towns supplied through public stand posts | 40 |
- These figures exclude unaccounted-for water; an allowance (commonly up to about 15%) is added for losses.
- Rural household tap connections under the Jal Jeevan Mission are planned for 55 lpcd.
Factors affecting per capita demand
Size of the city (larger cities — higher demand); climate (hot, dry — higher); living standards and habits of people; industrial and commercial activity; pressure in the distribution system (higher pressure — more wastage); quality of water; metering and cost (metering reduces demand); sewerage system (flush toilets increase demand); policy on continuous or intermittent supply; efficiency of the water works administration (leak control).
Fire demand
| Formula | Fire demand |
|---|---|
| Kuichling | (litres/min) |
| Freeman | (litres/min) |
| National Board of Fire Underwriters (central congested areas) | (litres/min) |
| Buston | (litres/min) |
| Indian practice (CPHEEO) | (kilolitres/day) |
Fire hydrants are provided at intervals along mains, and the distribution system (especially storage) is designed to meet fire flow at adequate pressure.
Variations in demand
Water demand varies:
- Seasonally — higher in summer.
- Daily — with days of the week and festivals.
- Hourly — two peaks in most towns (morning and evening), minimum at night.
- Maximum daily demand ≈ 1.8 × average daily demand
- Maximum hourly demand ≈ 1.5 × maximum daily demand = 2.7 × average daily demand
- Goodrich's formula: — percentage of the annual average demand for a period of days (e.g. about 180% for a day, 148% for a week)
Peak factors for distribution systems (CPHEEO):
| Population | Peak factor |
|---|---|
| Up to 50 000 | 3.0 |
| 50 000 to 2 lakh | 2.5 |
| Above 2 lakh | 2.0 |
| Rural water supply schemes | 3.0 |
Design flows for components
| Component | Designed for |
|---|---|
| Source, intake, raw water pumps and mains | Maximum daily demand (often average daily for sources with storage) |
| Treatment plant | Maximum daily demand |
| Service reservoirs | Hourly fluctuations (balancing), fire and emergency storage |
| Distribution system | Maximum hourly demand (or maximum daily + fire demand, whichever is greater) |
Design period
The design period is the number of years for which a component is designed to be adequate. It depends on the useful life of the component, ease of future expansion, rate of population growth, cost and interest rates, and the performance of the component in early years (under-loading).
Design periods recommended by CPHEEO (typical):
| Component | Design period (years) |
|---|---|
| Storage by dams | 50 |
| Intake structures, conveyance mains, distribution system, trunk mains | 30 |
| Pumping machinery | 15 |
| Water treatment units | 15 |
| Service (clear water) reservoirs | 15 |
| Land acquisition for future extensions | 30 |
The design period is counted from the expected year of completion of the project, not from the start of design.
Population forecasting
The future population is estimated from past census data (India's census is decennial).
1. Arithmetic increase method — constant increase per decade (old, large, saturated cities):
= average increase per decade.
2. Geometric increase method — constant percentage growth (young, rapidly growing cities):
= geometric mean of the decadal growth rates . Gives the highest estimate.
3. Incremental increase method — combines arithmetic and geometric trends (average cities):
= average of the increments (change in decadal increases).
4. Decreasing rate of growth method — the percentage growth rate decreases steadily; the average decrease in rate is subtracted from the latest rate each decade.
5. Logistic curve (S-curve) method:
using three census populations , , at equal time intervals; = saturation population.
Other methods:
- Simple graphical method — extend the population–time curve by eye.
- Comparative graphical method — the city's growth is assumed to follow that of larger, similar cities when they were of the same population.
- Master plan / zoning method — population based on planned densities of zones in the city's master plan.
- Ratio and correlation method — city population as a ratio of the state or national population forecast.
Arithmetic increase generally gives low estimates, geometric increase high estimates, and incremental increase intermediate values.
Worked examples
Census populations of a town are 40 000 (1981), 50 000 (1991), 62 000 (2001) and 76 000 (2011). Estimate the 2041 population by the arithmetic, geometric and incremental increase methods.
Solution. Decadal increases: 10 000, 12 000, 14 000 → = 12 000; increments: 2000, 2000 → = 2000; = 3.
Arithmetic:
Incremental:
Geometric: rates 25%, 24%, 22.58% → (approx.)
Find the fire demand for a city of 1 lakh population by Kuichling's and Freeman's formulas and the CPHEEO formula.
Solution. = 100 (thousands) Kuichling: Freeman: CPHEEO:
A city with sewerage has a design population of 1 lakh. Find the average daily demand, maximum daily demand and maximum hourly demand (use 135 lpcd).
Solution. Average L/day = 13.5 MLD Maximum daily Maximum hourly rate (as a rate)
Frequently tested points
- CPHEEO: 70 lpcd (no sewerage), 135 lpcd (with sewerage), 150 lpcd (metro), 40 lpcd (stand posts); JJM rural 55 lpcd.
- Metering and lower pressure reduce per capita demand; flush toilets increase it.
- Kuichling ; Freeman ; CPHEEO kL/day.
- Max daily = 1.8 × average; max hourly = 2.7 × average; Goodrich .
- Peak factor: 3.0 (≤ 50 000), 2.5 (50 000–2 lakh), 2.0 (> 2 lakh).
- Design periods: dams 50 y; intake, mains, distribution 30 y; pumps, treatment plants, service reservoirs 15 y.
- Arithmetic (old cities, lowest), geometric (young cities, highest), incremental (intermediate).
- Treatment plant designed for maximum daily demand; distribution for maximum hourly demand.
- Using the arithmetic mean of growth rates in the geometric method (use the geometric mean).
- Taking in years instead of decades.
- Designing distribution pipes for average daily demand.
- Water demand combines domestic, industrial, commercial, public and fire uses plus losses.
- CPHEEO norms and local factors fix per capita demand.
- Demand varies seasonally, daily and hourly; peak factors size different components.
- Design periods depend on component life and expandability.
- Population is forecast by arithmetic, geometric, incremental, logistic and graphical methods.