Water distribution systems have a very long and rich history dating back to the third millennium B.C. Advances in water supply and distribution were followed in parallel by discoveries and inventions in other related fields. Therefore, it is the aim of this paper to review both the history of water distribution systems and those related fields in order to present a coherent summary of the complex multi-stranded discipline of water engineering. Related fields reviewed in this paper include devices for raising water and water pumps, water quality and water treatment, hydraulics, network analysis, and optimisation of water distribution systems. The review is brief and concise and allows the reader to quickly gain an understanding of the history and advancements of water distribution systems and analysis. Furthermore, the paper gives details of other existing publications where more information can be found.

INTRODUCTION

Water distribution systems (WDSs) have an interesting and ancient history, which consists of not only steady advancements, but also deep declines. Undeniably, developments in WDSs and other engineering sciences emerged with the progress of nations, and stagnated or even retrogressed when nations declined. Ewbank (1858) states that ‘science and the arts are renovating the constitution of society [ … ]. Historians will hereafter trace in them the rise and fall of nations; for power and pre-eminence will depend upon new discoveries in and applications of science.’ Correspondingly, understanding the history of a particular engineering science is important in order to realise the broader linkages by which the science developed. This paper presents a coherent history of such developments in WDSs inclusive of devices for raising water and water pumps, water quality and water treatment, hydraulics, network analysis, and optimisation of WDSs.

Water distribution systems

The history of water distribution is very ancient in development. Indeed, urban WDSs date back to the Bronze Age (circa 3200–1100 B.C.), with ‘several astonishing examples’ from the mid-third millennium B.C. (Mays et al. 2012). These include, for example, a system of hundreds of wells supplying water to domestic demands, and private and public baths (Mays et al. 2012). Crouch (1993), who documented water management in ancient Greece, revealed that the very first piped water supplies including pressure pipes had been known as early as the second millennium B.C. It is documented that ancient Minoan and Greek civilisations had urban water reticulation, sewerage and drainage systems, with wells, cisterns, tanks, reservoirs, dams, channels and water pipes made of terracotta (clay) and lead (Crouch 1993; Angelakis et al. 2005; Mays et al. 2012). Moreover, the ancient Greeks constructed ‘long-distance water supply lines with tunnels and bridges’ referred to as aqueducts, which are dated back to the eighth to sixth century B.C. (Crouch 1993).

Greek technologies were subsequently inherited by the Romans (circa 100 B.C. to 500 A.D.), who developed them further and implemented them at an enlarged scale (Mays et al. 2007; Angelakis et al. 2012). In particular, Roman aqueducts, which carried water from a source to the Roman cities, could extend over more than 100 km in length (Viollet 2000; Haut & Vivier 2012). They could incorporate an inverted siphon, which was a pressurised pipeline carrying water across a valley (Haut & Vivier 2012). The Romans also used wooden pipes as an alternative to the terracotta pipes, prevalent in Northern Europe (Hodge 2002). The durability of the Roman constructions is remarkable, with some of them having operated up to modern times (Mays et al. 2012). Furthermore, it is recognised that the Romans had an advanced knowledge of water supply engineering (Hodge 2002; Haut & Vivier 2012).

After the fall of the Roman Empire at circa fifth century A.D., it is unclear if the Roman knowledge about water management survived the collapse of these civilisations (Crouch 1993) and the following Middle Age period referred to as Dark Ages (5th–15th centuries A.D.). Even though it is agreed that Roman achievements ‘were not totally forgotten’ (Angelakis et al. 2012), it is admitted that there was a decline in the quality of water management practices during those several centuries (Burian & Edwards 2002; Angelakis et al. 2012). This decline with very poor sanitary conditions including polluted water in sources and waste in the streets is reported, especially in Europe (Gray 1940). Water supply was provided to a central delivery point, from where it was brought to the homes by either people themselves or servants, or else water carriers who made a business of selling and delivering water (Figure 1).

Figure 1

Water carrier, carrying water in so called tankards (Smiles (1862) and cited by Draffin (1939)).

Figure 1

Water carrier, carrying water in so called tankards (Smiles (1862) and cited by Draffin (1939)).

It was not until after the Renaissance (14th–17th centuries) when water management practices began to evolve once more (Walski 2006). Possibly the first major pipeline was a 25 km line from Marly-on-Seine to the Palace of Versailles in France, which was completed in 1664 (Walski 2006). By the mid-1700s, London had more than 50 km of water mains constructed of wood, cast-iron and lead pipes (Sanks 2005; Walski 2006). In the United States, the first piped water supply was in Boston in 1652, when water was brought from springs and wells to near what is now the restored Quincy Market area (Mays 2000).

More information about the history and evolution of water supply can be found in Angelakis et al. (2012), Draffin (1939) and Viollet (2000).

Devices for raising water and water pumps

It is believed that ancient devices for raising water originate from the Assyrians, Babylonians and other ancient nations (Ewbank 1858). The early devices used to raise water from wells are a pulley bucket, windlass and their various modifications to transfer motion. While they may seem basic to mention here, they were of a great significance. For example, the theft of a ‘fatal bucket’ from a public well in Bologna, by Modena soldiers, is believed to have resulted in a war between those two nations (the incident occurred at the beginning of the 11th century A.D.) (Ewbank 1858). Other early devices were invented for agricultural purposes to raise water over small elevations, such as a (swinging) gutter and a combination of levers and gutters. The most wide spread, nevertheless, appears to be the swape also known as the shadoof or the ‘counterbalanced bailing bucket’ (Needham 1965), which had been in use since the second millennium B.C. and was still very common on the European continent in the 18th century A.D. (Ewbank 1858).

In the first century B.C., Vitruvius compiled the existing knowledge of hydraulics (Rouse & Ince 1963) and described the principal hydraulics mechanisms to raise water invented since antiquity to date (Ewbank 1858; Pollio et al. 1859). These included the noria, the tympanum, the chain of pots, also referred to as sakia, the screw pump and the machine of Ctesibius of Alexandria (the force pump). The screw and force pumps are often reported as the first pumps (in terms of modern meaning) in the literature (Rouse & Ince 1963; Crouch 1993; Koutsoyiannis & Angelakis 2003); the force pump was the first pump discharging water by pressure exerted artificially, not by gravity (Viollet 2000).

The tympanum and the noria (Egyptian ‘wheel of fortune’) both have the form of a wheel partially submerged into water, in which water is elevated by gutters or vessels, respectively, using its rotary motion (Ewbank 1858). Interestingly, the tympanum was improved centuries later by De La Faye, a member of the French Royal Academy of Sciences, to convey water via spirals (De Bélidor & Navier 1819). Although the tympanum and noria are limited in lifting water by the diameter of the wheel, Needham (1965) reports a c. 50 ft (15 m) noria made of bamboo and wood on the Yellow River in China, and ‘the most splendid’ c. 70 ft (21 m) noria on the Orontes in Syria. The chain of pots (Figure 2), which consists of endless rope carrying vessels at equal distances, which fill at the bottom and discharge at the top (Ewbank 1858; Needham 1965), was invented to lift water from ‘every depth’ (Ewbank 1858). The most remarkable application is Joseph's well at Cairo consisting of two consecutive wells of total depth 297 ft (91 m), which are divided by an animal driven chamber at 165 ft (50 m) (Ewbank 1858). According to Ewbank (1858), the chain of pots, often driven by windmills ‘constituted “water works” for supplying European cities’ previous to the 16th century, and was still employed throughout Asia and Europe in the 19th century.

Figure 2

(a) A chain of pots in Spain, (b) upper section of a modern chain of pots (Ewbank 1858).

Figure 2

(a) A chain of pots in Spain, (b) upper section of a modern chain of pots (Ewbank 1858).

The invention of the most original screw pump, which is used today in stormwater and wastewater applications, is attributed to Archimedes (circa 287–212 B.C.) (Koutsoyiannis & Angelakis 2003). The force pump nowadays called the reciprocating pump, which consists ‘essentially of two piston pumps with the discharge pipes connected together through an air chamber’, is attributed to Ctesibius of Alexandria (circa 285–222 B.C.) (Draffin 1939). Vitruvius refers to this pump as ‘the machine of Ctesibius which expels water to a great height’ (Blackstone 1957), as it was used, for example, to pump water from wells or create water jets to fight fires (Koutsoyiannis & Angelakis 2003). Pieces of force pumps were discovered in Hampshire, the site of the Roman town Calleva Atrebatum, and are on display at the British Museum (Blackstone 1957).

Those pumps were powered manually either by humans or animals, noria when drawing water from a river could also be powered by water current (Needham 1965), and the chain of pots was later often driven by windmills (Ewbank 1858). Even though the power of steam had been known in antiquity through the invention of aeolipile by Heron of Alexandria (circa 10–70 A.D. (Wikipedia 2014)), it did not find any industry application (Viollet 2000) until the invention of the steam engine in the 17th century. The predecessors of steam pumps are described in the literature as ‘machines for raising water by (aid of) fire’ (Ewbank 1858).

Although the principle of suction was known since antiquity and suction pumps were used for a long time, it was not until 1643, when an Italian, Evangelista Torricelli explained it (Draffin 1939). Though Torricelli's explanation was persistently denied by certain groups, the principle became well established through the subsequent experiments on air pressure by Boyle and Guericke, the latter of whom invented the air pump (Draffin 1939). A significant improvement in pump development was the rotary pump, plans for which were published by an Italian, Ramelli, in 1588 and the centrifugal pump (Draffin 1939; Ewbank 1858) invented in the late 17th century.

The first large pumped water supply system, of which there are adequate records, is the water supply system in London from the late 16th century. The system sourced water from the River Thames, with pumps installed under London Bridge in 1582 (Draffin 1939). The most spectacular pump installation was on the Seine to supply the Palace of Versailles in France, which was constructed in 1682. The pumps were arranged into three levels and raised water subsequently into three reservoirs, the highest of which was located 533 ft (162 m) above the Seine and 1.2 km from it. Pumps were driven by wooden water wheels, which were later replaced with a 64-horsepower steam engine (Draffin 1939).

More information about the history and evolution of devices for raising water and water pumps can be found in Draffin (1939), Ewbank (1858) and Needham (1965).

Water quality and water treatment

‘Clean and safe water is the most important consideration of a healthy population, community, and economy’ (Pope et al. 2012). Mays et al. (2012) states that historically, drinking water has been considered clear water; hence, the first treatment attempts were aiming at the improvement of its aesthetic conditions. The earliest recorded knowledge of water treatment comes from Indian sources dated circa 2000 B.C. One of them suggests that ‘it is good to keep water in copper vessels, to expose it to sunlight, and filter through charcoal’, another directs ‘ … to heat foul water by boiling and exposing to sunlight and by dipping seven times into it a piece of hot copper, then to filter and cool in an earthen (terracotta) vessel’ (Baker 1949).

The first development of water treatment technology relevant to urban supply lies in Minoan civilisation at the beginning of the Bronze Age (circa 3200–1100 B.C.). In those days, sedimentation cisterns were used for the removal of suspended solids, and terracotta infiltration devices filled with charcoal for removing both organic and inorganic constituents (Sklivaniotis & Angelakis 2006; Mays et al. 2012). In the sixth century B.C. at the latest, utilisation of definitely two and probably three qualities of water, potable (i.e. springs), sub-potable (i.e. cisterns) and non-potable (i.e. storm runoff probably combined with waste waters for irrigation) is documented (Crouch 1993), which indicates a very high standard of water management practices. This agrees well with a discussion by the Greek physician Hippocrates (460–377 B.C.), the father of medicine, in relation to the qualities of water sources, who recommended to select ‘the most health-giving sources of supply rather than (on) rectifying the waters that were bad’ (Baker 1949). He also introduced a cloth bag later known as ‘Hippocrates' Sleeve’ for straining rain water which he suggested to be previously boiled (Baker 1949). Greek historian Herodotus (fifth century B.C.) stated that water drunk by the Persian kings was previously boiled and kept in silver vessels (Ewbank 1858). The Romans (circa 100 B.C.–500 A.D.) frequently boiled rainwater before they used it (Ewbank 1858).

To bring water from a source to the cities, Greeks and Romans constructed aqueducts typically consisting of open channels, tunnels and bridges. In the first century A.D., Frontinus wrote a treatise describing Roman water supply methods (Rouse & Ince 1963), a particular interest being his ‘description of a settling reservoir at the head of one of the aqueducts supplying Rome and of … ingeniously designed pebble catchers built into most of the aqueducts’ (Baker 1949). Water from aqueducts was delivered to houses, where it was stored in cisterns for use by families (Crouch 1993). According to Ewbank (1858), if water was required to be perfectly pure, two or three cisterns were built at different levels, ‘so that the water successively deposited the impurities’.

Similar to other scientific fields, water treatment technologies had not progressed during the Middle Ages, often referred to as the Dark Ages, following the fall of the Roman Empire in about the fifth century A.D. (Mays et al. 2012). Discoveries started emerging at the beginning of the 17th century. In 1627, Francis Bacon's experiments dealing with water purification, such as percolation, filtration, distillation and coagulation, were published (Mays et al. 2012). The important invention of the microscope, which led subsequently to a new field of bacteriology, dates back to the first decade of the 17th century, when Galileo Galilei developed the compound microscope to observe insects. Microscopes were subsequently popularised in Hooke's books from 1665 and 1678, the second of which gives detailed instructions for making microscopes (Wilson 1995). In the late 17th century, Antony van Leeuwenhoek developed a microscope powerful enough to see bacteria (Wilson 1995). In 1675, he discovered ‘living creatures’ in the water which were ‘continually moving themselves’ and called them ‘animalcula’ (Baker 1949). In spite of those bacteriological discoveries, which were originally considered ‘as unimportant curiosities’, it took another 200 years to understand their importance (Mays et al. 2012).

It was in 1849, when John Snow related cholera to public water supplies, and in 1854, when he investigated a cause of cholera outbreak in London as being a central water supply point, the Broad Street pump/well (Snow 1849; Snow 1855). However, ‘Snow's work was not accepted by the medical establishment’ at that time (Savic & Banyard 2011). Although the handle of the pump was removed by the public authorities and water was not used during the outbreak, the handle was reinstalled once the epidemic had passed (Savic & Banyard 2011). About 30 years later, the impact of bacteria in drinking water on human health was eventually recognised when the French scientist Louis Pasteur proved the ‘germ theory of disease’ correct and the more widely accepted miasmic theory invalid (Savic & Banyard 2011).

In the 17th century, the patent era in water treatment commenced. The first known illustrated description of sand filters was published by Luc Antonio Porzio in the late 17th century (Baker 1949). In the 18th century, the first patent for a water filter made of sponge, wool and sand was granted to Joseph Amy (Amy 1754), and James Peacock received a patent for a sand filter with backwashing (Mays 2013). According to Baker (1949), the first filtration plant to supply a whole town was completed at Paisley, Scotland in 1804. In 1806, a large water treatment plant (WTP) opened in Paris (Mays et al. 2012), which sourced water from the River Seine and was continually operated for a half century (Baker 1949). It used gravity filters ‘modelled on the Smith-Cuchet-Montfort patent of 1800’ and composed of layers of different sand fractions and pounded charcoal (Baker 1949). In 1829, James Simpson's slow sand filters were completed in London and later ‘became the model for English slow sand filters throughout the world’ (Baker 1949). In 1838, Theophile Ducommun obtained a French patent for a lateral-flow pressure filter and in 1835, Louis-Charles-Henri de Fonvielle for a high-pressure filter (Baker 1949). The first rapid sand filters were applied in New Jersey in 1882 (Mays et al. 2012). In 1895, Allen Hazen ‘wrote the first treatise on the art and science of water filtration’ (Baker 1949).

The earliest proposals to disinfect water were made before there was knowledge of waterborne diseases. The first of these found on record is a statement by Dr Robley Dunglinson published at Philadelphia in 1835 (Baker 1949). One of the first known uses of chlorine for water disinfection was by John Snow in 1850, when he attempted to disinfect the Broad Street well in London after an outbreak of cholera (Christman 1998). In 1906, ozone was used for the first time for disinfection in Nice, France and became very popular in Europe, whereas chlorine was mostly used for disinfection in the USA (Mays et al. 2012). The first serious effort in water desalination was undertaken during World War II for units which encountered difficulties in securing drinking water (Mays et al. 2012).

More information about the history and evolution of water treatment can be found in Baker (1949), Draffin (1939) and Mays (2013).

Hydraulics

The field of hydraulics has evolved through millennia. Basic hydraulic principles that ‘water flows downhill’ and ‘water always raises to its own level’ had been applied by ancient civilisations (Hodge 2002) to construct irrigation and water supply systems several millennia ago. In those days, hydraulics was ‘purely an art, with no scientific basis beyond the principle’ (Rouse & Ince 1963) and the complexity of designed water systems surpassed the ability to mathematically describe such phenomena (Crouch 1993).

The admirable pioneers in the field of hydraulics were Archimedes (circa 287–212 B.C.), who established the principles of buoyancy (Pollio et al. 1859) and Heron of Alexandria (circa 10–70 A.D. (Wikipedia 2014)), who was the first to formulate the relationship between flow, velocity and cross-sectional area (Viollet 2000; Walski 2006). In the first century B.C., Vitruvius compiled the existing knowledge on hydraulics and subsequently in the first century A.D., Frontinus wrote a treatise describing Roman water supply methods (Rouse & Ince 1963). Koutsoyiannis & Angelakis (2003) report that unfortunately many of those early hydraulic theories were forgotten for centuries of the Middle Ages to come, ‘only to be re-invented during the Renaissance or later’.

The Renaissance represented a perceptible change from the philosophical to observational science and hydraulics started progressively evolving based on experimental or empirical approaches (Rouse & Ince 1963). In the 15th century, Leonardo da Vinci made observations of many flow phenomena and expressed an elementary principle of continuity using the ‘analogy of a dense crowd having to move with increasing speed through a passage of decreasing width’ (Rouse & Ince 1963). Interestingly, da Vinci's continuity principle did not appear to be widely spread at that time, but it was eventually rediscovered and popularised by Castelli in the early 17th century (Rouse & Ince 1963). In the 18th century, Newton introduced important laws of motion, which subsequently allowed an understanding of hydraulics, and Bernoulli developed principles of fluid flow (Walski 2006). In 1883, laminar and turbulent flow was defined by Reynolds (Reynolds 1883).

The development of an important headloss equation, which predicts headloss for pipe or channel flow, represents over two centuries of incremental discoveries. In the 1770s, Chezy formulated the first headloss equation, which was extended by Darcy and Weisbach to a more general form in 1845 (Brown 2002; Walski 2006). In 1891, Manning proposed headloss equations for flow in open channels and pipes (Manning et al. 1891). In 1906, Hazen and Williams developed a headloss equation using ‘C-factor’ rather than friction (Walski 2006). In 1944, Moody related friction factor, roughness and Reynolds number (Moody & Princeton 1944), and more recently, Allen related the Hazen–Williams and Darcy–Weisbach headloss equations (Allen 1996). Nowadays, the most commonly used equations for pressurised pipe networks are the Hazen–Williams and Darcy–Weisbach equations. Savic & Walters (1997) made a valuable observation about the Hazen–Williams equation in relation to its different interpretations, which led to inconsistencies in network performance predictions.

The field of hydraulics recognises much more famous names and discoveries, which are outside the scope of this paper and can be found in Rouse & Ince (1963).

Network analysis

Network analysis, which is invaluable for the water professional involved with design, operation, maintenance and optimisation of WDSs, consists of two distinct components, namely, analyses of (i) hydraulic and (ii) water quality behaviour of flow through a WDS. This section focuses on hydraulic analysis only. Hydraulic analysis calculates flows, headlosses and pressures in a specified pipe network by simultaneously solving a set of equations (further in the text referred to as ‘network equations'). These network equations arise from the conservation of mass of flow and energy as (i) the sum of flows toward any junction is zero, (ii) the sum of headlosses in a closed loop is zero and (iii) the headloss in a pipe is directly proportional to the power of the flow (Aldrich 1937). Due to the size of WDSs and the associated large number of nonlinear network equations, hydraulic analysis is a complex task. Historically, hydraulic analysis methods range from graphical methods, through the use of physical analogies, to mathematical models (Ramalingam et al. 2004).

Prior digital computers

Hydraulic analysis of WDSs involves tedious calculations applying a combination of simplifications, engineering experience and practice, and conservatism (Walski et al. 2006). The first method reported was the graphical method introduced by Spiess (1887) and (Aldrich 1937), followed by the more popular graphical method of Freeman (1892). The former method presented solutions for basic branched and looped systems, whereas the latter method investigated simple and more complex WDSs with fire demands. Freeman's graphical method was later expanded by Aldrich (1937) using the Hazen–Williams formula. Other well-known methods include the electric network analyser method (Camp & Hazen 1934) based on the analogy between the laws governing hydraulic flow and electric current in networks (Ramalingam et al. 2004) and the Hardy–Cross method (Cross 1936), which was the first method to solve hydraulic analysis mathematically. The graphical and electric analyser methods have not been widely used due to time and equipment requirements, respectively (Aldrich 1937), the Hardy–Cross method became popular with numerous subsequent publications describing its application to various systems (Ramalingam et al. 2004).

After digital computers

Several iterative methods have been applied to hydraulic analysis of a WDS. The first method adapted to the digital computer was the Hardy–Cross method (Cross 1936) in 1957, with application to the WDS of the city of Palo Alto, California (Ormsbee 2006). Because this method could take a long time to converge to a solution or could fail to converge at all, other methods were proposed (Ormsbee 2006). These methods included the Newton–Raphson method (simultaneous node method) (Martin & Peters 1963), simultaneous loop method (Epp & Fowler 1970), the linear theory approach (simultaneous pipe method) (Wood & Charles 1972; Tavallaee 1974) and the gradient method (simultaneous network method) (Todini & Pilati 1988). The Newton–Raphson method may converge more quickly for small networks, but very slowly for large networks compared to the linear theory approach (Mays 1989). The simultaneous loop method is the improved Newton–Raphson method with the benefit of significantly improved convergence characteristics of the original algorithm (Ormsbee 2006). The linear theory approach has the capacity to analyse all network components and is more flexible regarding the representation of pumps (Mays 1989). The gradient method was adopted in the development of the hydraulic simulation package EPANET (Rossman 1993).

The next significant step in hydraulic analysis of WDSs was development of hydraulic simulation packages, accessible for wide use by water professionals. The first such package titled KYPIPE, which uses the simultaneous loop method to solve the network equations, was developed by the University of Kentucky in 1980 (Wood 1980). Another package, WADISO, has been introduced by the US Army Engineer Waterways Experiment Station and uses the simultaneous node method to solve the network equations. KYPIPE and WADISO are compared in Mays (1989). Possibly the most widely spread has become the simulation software EPANET (Rossman 1993), which is used in other (commercial) hydraulic analysis packages. Those hydraulic simulation packages have become well accepted tools (Mays 1989; Van Dijk et al. 2008) and are used nowadays in conjunction with optimisation techniques to solve optimum WDS design, operation and other related optimisation problems in WDSs.

More information about network analysis can be found in Camp (1943), Ormsbee (2006) and Ramalingam et al. (2004).

WDS optimisation

There are at least a dozen literature review papers on optimisation of WDSs which have been published since the 1970s until nowadays (Table 1). These papers review mainly publications since the 1960s/1970s to date; some of them also reference the publication of Camp (1939) as the first work in the field. However, even older publications can be found, with the oldest dating back to the 1890s. Hence, it appears that the formal work in optimisation of WDSs commenced about half a century before it is commonly reported. The following section addresses this gap by reviewing the publications from 1890s to the 1950s to link to the existing literature review papers listed in Table 1.

Table 1

Publications containing literature review of Water distribution system optimisation in chronological order

Author(s), yearPaper titleDate of referencesComment
Shamir (1974)  Optimal design and operation of water distribution systems 1961–1972 Review of optimisation of WDS design (no work was found to date concerning optimisation of WDS operation) 
Shamir (1979)  Optimisation in water distribution systems engineering 1963–1977 Review of optimisation of WDS design and operation, including mathematical formulations of optimisation models 
Walski (1985)  State-of-the-art pipe network optimisation 1931–1939, 1968–1985 Classifies papers into categories of fixed flow pattern (branched systems), variable flow pattern (looped systems), gravity and pumped systems 
Lansey & Mays (1989)  Optimisation models for design of water distribution systems 1939, 1961–1988 Apart from the literature review includes also general optimisation model (cost of pipes, pumps and storage), and WDS design for multiple loading conditions (solution methodology and application) 
Walters (1992)  A review of pipe network optimisation techniques 1966–1991 Review of optimisation techniques for WDS design, which is divided into branched and looped networks. Also includes discussion on reliability 
Dandy et al. (1993)  A review of pipe network optimisation techniques 1936, 1963–1992 Review and comparison of four optimisation techniques: partial enumeration, nonlinear programming (NLP), linear programming (LP), genetic algorithms (GA) 
Ostfeld & Shamir (1993)  Incorporating reliability in optimal design of water distribution networks – review and new concepts 1972–1992 Paper contains (i) discussion on definition of reliability, (ii) review of optimisation techniques for WDS design to include reliability, (iii) new concept for incorporation of reliability into optimal WDS design 
Ormsbee & Lansey (1994)  Optimal control of water supply pumping system 1968–1994 Review of optimisation techniques used for a pump scheduling problem, inclusive of detailed review table 
Simpson et al. (1994)  Genetic algorithms compared to other techniques for pipe optimisation 1973–1992 Overview of deterministic techniques (enumeration, nonlinear programming) and GA for WDS design. Also includes application of GA and its comparison with the two above deterministic techniques 
Engelhardt et al. (2000)  Rehabilitation strategies for water distribution networks: a literature review with a UK perspective 1972–1999 Review of optimisation models for WDS rehabilitation, inclusive of models for extended planning horizons and multi-objective optimisation approaches 
Lansey (2006)  The evolution of optimising water distribution system applications 1939, 1961–2006 Review of WDS optimisation with chronological–topical charts, chronological–statistical charts. Also outlines future needs in the field 
Nicklow et al. (2010)  State of the art for genetic algorithms and beyond in water resources planning and management 1969–2007 (WDS optimisation only) Review of applications of evolutionary algorithms (EAs) in water resources planning and management, including WDS optimisation. Future challenges are highlighted 
Author(s), yearPaper titleDate of referencesComment
Shamir (1974)  Optimal design and operation of water distribution systems 1961–1972 Review of optimisation of WDS design (no work was found to date concerning optimisation of WDS operation) 
Shamir (1979)  Optimisation in water distribution systems engineering 1963–1977 Review of optimisation of WDS design and operation, including mathematical formulations of optimisation models 
Walski (1985)  State-of-the-art pipe network optimisation 1931–1939, 1968–1985 Classifies papers into categories of fixed flow pattern (branched systems), variable flow pattern (looped systems), gravity and pumped systems 
Lansey & Mays (1989)  Optimisation models for design of water distribution systems 1939, 1961–1988 Apart from the literature review includes also general optimisation model (cost of pipes, pumps and storage), and WDS design for multiple loading conditions (solution methodology and application) 
Walters (1992)  A review of pipe network optimisation techniques 1966–1991 Review of optimisation techniques for WDS design, which is divided into branched and looped networks. Also includes discussion on reliability 
Dandy et al. (1993)  A review of pipe network optimisation techniques 1936, 1963–1992 Review and comparison of four optimisation techniques: partial enumeration, nonlinear programming (NLP), linear programming (LP), genetic algorithms (GA) 
Ostfeld & Shamir (1993)  Incorporating reliability in optimal design of water distribution networks – review and new concepts 1972–1992 Paper contains (i) discussion on definition of reliability, (ii) review of optimisation techniques for WDS design to include reliability, (iii) new concept for incorporation of reliability into optimal WDS design 
Ormsbee & Lansey (1994)  Optimal control of water supply pumping system 1968–1994 Review of optimisation techniques used for a pump scheduling problem, inclusive of detailed review table 
Simpson et al. (1994)  Genetic algorithms compared to other techniques for pipe optimisation 1973–1992 Overview of deterministic techniques (enumeration, nonlinear programming) and GA for WDS design. Also includes application of GA and its comparison with the two above deterministic techniques 
Engelhardt et al. (2000)  Rehabilitation strategies for water distribution networks: a literature review with a UK perspective 1972–1999 Review of optimisation models for WDS rehabilitation, inclusive of models for extended planning horizons and multi-objective optimisation approaches 
Lansey (2006)  The evolution of optimising water distribution system applications 1939, 1961–2006 Review of WDS optimisation with chronological–topical charts, chronological–statistical charts. Also outlines future needs in the field 
Nicklow et al. (2010)  State of the art for genetic algorithms and beyond in water resources planning and management 1969–2007 (WDS optimisation only) Review of applications of evolutionary algorithms (EAs) in water resources planning and management, including WDS optimisation. Future challenges are highlighted 

The early publications on WDS optimisation do not refer to system ‘optimisation’, but rather system ‘economy’. They obtain optimum solution (i.e. a minimum of function) by placing the first derivate equal to zero. This approach was used priorly in the field of operations research with mathematical optimisation, which was established during World War II, when there was a need to resolve strategic and tactical problems using limited military resources (Taha 1992).

The first written record found concerning the economic aspects of water works are from the meetings of the New England Water Works Association (NEWWA) from the late 19th century. Nevons, the NEWWA president at the time, refers to economy of water works as ‘the subjects […] of the greatest importance’ (Nevons 1889), and Allis (1892) talks about the importance of including more discussions on economic aspects of water works in the NEWWA meetings.

As early as 1895, Tuttle presented the work for economic pipe sizes in WDSs by using economic velocity of flow through pipes (Tuttle 1895). Being possibly the first work, it is described here in more detail. Tuttle based the paper on the knowledge that the decrease in pipe sizes, and consequently costs of the pipes, increases headlosses and thus the pressure required, and vice versa. He formulated an equation representing the annual cost of a WDS including initial capital cost, annual interest plus depreciation and annual operating cost for pumping. Placing the derivative of this equation equal to zero, he minimised pipe diameter and subsequently calculated the ‘economic velocity.’ The results were summarised in a tabular form for a range of pipe diameters. Tuttle's approach included several assumptions, such as the cost of cast-iron pipes, pipelaying, pumping and others, which were implemented as constants. On the other hand, he introduced a factor for variable demand.

The principle of economic velocity to determine economic pipe sizes was later used in the work of True (1937) and Braca & Happel (1953). Some advancements in True's work were, for example, implementation of a range of Hazen–Williams coefficients for two pipe materials, cast-iron and steel, and use of the variable cost of pipelaying. He also stated the importance of ‘proper engineering allowances’ made for future increase in water consumption. Braca & Happel (1953) utilised two methodologies to determine economic pipe sizes, first the principle of economic velocity and second the minimum annual costs for pipes and pumps which he divided into initial capital costs and ongoing costs.

In the publications of Genereaux (1937b), Camp (1939), Lischer (1948) and Sarchet & Colburn (1940), nonetheless, the principle of economic velocity was abandoned due to its inaccuracies. Genereaux (,1937a, b) included pipe and pumping costs in the total annual costs and developed convenient charts, from which economic pipe diameters could be derived with a limitation to turbulent flows only. His work was later extended by Sarchet & Colburn (,1940, 1941) for both the turbulent and viscous flows. Their results were also presented in practical charts for determination of economic pipe diameters. Camp (1939) expressed the total annual cost of a WDS as the sum of initial capital and labour costs, annual interest plus depreciation and annual operating cost for pumping. Camp's work received a lot of interest and was commented on by numerous authors (Camp 1939). The discussion topics were, for example, how to determine exact prices of pipes, pipe laying and pumping, how to predict pipe roughnesses over time and (maximum) future demands and demand patterns. Lischer (1948) applied Camp's principles of design to a simple system with a pump and storage tank and showed interrelations between the variables.

The determination of the most economic pipe sizes were subject to early research not only in WDSs, but also in high-pressure water-power installations (Adams 1907; Jeffcott 1928). These studies were based on the proposition that the annual cost of the pipes plus the annual value of the energy sacrificed due to friction is a minimum. Jeffcott (1928) proposed a numerical solution considering both pipes and tunnels excavated through rocks, while Adams (1907) developed a graphical solution. Adams' work was subsequently discussed by several authors (Butcher et al. 1907), some of whom verified the graphical solution mathematically. Additionally, it was pointed out that the previously mentioned proposition is a modification of a law in the electrical transmission of energy regarding the most economical area of a conductor, first proposed by Sir William Thomson in 1881 (Butcher et al. 1907).

Future research and development

This section is concerned with optimisation of WDSs within a rapidly growing field. There has been a dramatic increase in the development and application of evolutionary algorithms (EAs) over the last two decades (Nicklow et al. 2010) with future challenges and directions being actively proposed by authors (Nicklow et al. 2010; Maier et al. 2014). Nicklow et al. (2010) briefly summarise future needs as solving large-scale problems, advancing adaptive decision making under uncertainty, and bridging process science and operational engineering management. Conversely, Maier et al. (2014) analyse the development and application of EAs in depth and introduce an elaborate road map of future research challenges and directions. These challenges include development of improved problem formulation to account for uncertainty, understanding the impact of assumptions and simplifications, development of methods for search space reduction, methods for parameter selection to adapt during optimisation runs, improved performance measures for multi-objective problems, improved visualisation and communication tools to support decision making and others. Addressing these challenges will enable EAs to be applied to real-world problems with efficiency and confidence (Maier et al. 2014).

CONCLUSION

This paper presented a coherent picture of the history and advancements of WDSs and other related fields since the third millennium B.C. The related fields reviewed include devices for raising water and water pumps, water quality and water treatment, hydraulics, network analysis, and optimisation of WDSs. Concerning WDS optimisation, the reviewed publications cover a period from the 1890s to 1950s, which bridges the gap in the existing review literature. Future research directions are also incorporated.

The value of this paper is in bringing together a complex array of work allowing the reader to quickly gain an understanding of the history and advancements of WDSs and analysis. Moreover, the paper gives details of other existing publications, listed at the end of each section, where more information can be found.

ACKNOWLEDGEMENTS

The authors wish to acknowledge library personnel of the Federation University Australia (FedUni), namely Angela Thomas and Donna Byrne, for delivering references for this paper. Your persistent and professional approach, Angela and Donna, in obtaining very old and incorrectly cited references by sources was exceptional and we sincerely thank you. FedUni is very fortunate to have you in their library team. This work was supported by the Australian Research Council as Project LP0990908.

REFERENCES

REFERENCES
Adams
A. L.
1907
A solution of the problem of determining the economic size of pipe for high-pressure water-power installation
.
Transactions of the American Society of Civil Engineers
59
(
2
),
173
177
.
Aldrich
E. H.
1937
Solution of transmission problems of a water system
.
Transactions of the American Society of Civil Engineers
103
(
1
),
1579
1619
.
Allis
S. M.
1892
Economy in water-works management
.
Journal of New England Water Works Association (JNEWWA)
6
(
4
),
161
163
.
Amy
J.
1754
Suite du Livre Intitulé Nouvelles Fontaines Filtrantes, Approuvées par l'Académie Royale des Sciences
.
Antoine Boudet
,
Paris, France
(
in French)
.
Angelakis
A. N.
Koutsoyiannis
D.
Tchobanoglous
G.
2005
Urban wastewater and stormwater technologies in Ancient Greece
.
Water Research
39
(
1
),
210
220
.
Angelakis
A. N.
Mays
L. W.
Koutsoyiannis
D.
Mamassis
N.
2012
Evolution of Water Supply Through the Millennia
.
IWA Publishing
,
London, UK
.
Baker
M. N.
1949
The Quest for Pure Water: The History of Water Purification from the Earliest Records to the Twentieth Century
.
New York, USA
.
Blackstone
G. V.
1957
A History of the British Fire Service
.
Routledge and Kegan Paul
,
London, UK
.
Braca
R. M.
Happel
J.
1953
New cost data bring economic pipe sizing up to date
.
Chemical Engineering
60
(
1
),
180
187
.
Brown
G. O.
2002
The history of the Darcy-Weisbach equation for pipe flow Resistance
. In:
150th Anniversary Conference of ASCE
(
Rogers
J.
Fredrich
A.
, eds).
American Society of Civil Engineers (ASCE)
,
Reston, VA, USA
, pp.
34
43
.
Burian
S. J.
Edwards
F. G.
2002
Historical perspectives of urban rainage
. In:
9th International Conference on Urban Drainage
,
Portland, Oregon
,
8–13 September 2002
.
Butcher
W. L.
Buck
W. E.
Stevens
P. E.
DeMott
C. L.
Kuichling
E.
Roberts
W. J.
Marx
C. D.
Wing
C. B.
Wise
J. H.
Myers
E. W.
1907
Discussion of ‘A Solution of the Problem of Determining the Economic Size of Pipe for High-Pressure Water-Power Installation’
.
Transactions of the American Society of Civil Engineers
59
(
2
),
178
194
.
Camp
T. R.
1939
Economic pipe sizes for water distribution systems
.
Transactions of the American Society of Civil Engineers
104
(
1
),
190
213
.
Camp
T. R.
1943
Hydraulics of distribution systems - some recent developments in methods of analysis
.
Journal of New England Water Works Association (JNEWWA)
42
,
334
368
.
Camp
T. R.
Hazen
H. L.
1934
Hydraulic analysis of water distribution systems by means of an electric network analyzer
.
Journal of New England Water Works Association (JNEWWA)
48
,
383
404
.
Christman
K.
1998
The history of chlorine
.
Water World
14
(
8
),
66
67
.
Cross
H.
1936
Analysis of flow in networks of conduits or conductors
.
Bulletin No. 286, University of Illinois Engineering Experiment Station
34
(
22
),
1
29
.
Crouch
D. P.
1993
Water Management in Ancient Greek Cities
.
Oxford University Press
,
New York, USA
.
Dandy
G. C.
Simpson
A. R.
Murphy
L. J.
1993
A review of pipe network optimisation techniques
. In:
Watercomp 93: 2nd Australasian Conference on Computing for the Water Industry Today and Tomorrow
.
Australian Institute of Engineers
,
Melbourne, Australia
,
30 March–1 April 1993
, pp.
373
383
.
De Bélidor
B. F.
Navier
C. L. M. H.
1819
Architecture Hydraulique: ou, L'art de Conduire, d’Élever, et de Ménager les Eaux pour les Différents Besoins de la Vie
.
F. Didot
,
Paris, France
(
in French)
.
Draffin
J. O.
1939
The Story of Man's Quest for Water
.
The Garrard Press
,
Champaign, IL
.
Engelhardt
M. O.
Skipworth
P. J.
Savic
D. A.
Saul
A. J.
Walters
G. A.
2000
Rehabilitation strategies for water distribution networks: a literature review with a UK perspective
.
Urban Water
2
(
2
),
153
170
.
Epp
R.
Fowler
A. G.
1970
Efficient code for steady-state flows in Networks
.
Journal of the Hydraulics Division
96
(
1
),
43
56
.
Ewbank
T.
1858
A Descriptive and Historical Account of Hydraulic and Other Machines for Raising Water
.
Derby & Jackson
,
New York, USA
.
Freeman
J. R.
1892
The arrangement of hydrants and water-pipes for the protection of a city against fire
.
Journal of New England Water Works Association (JNEWWA)
7
(
1
),
49
81
.
Genereaux
R. P.
1937a
Correction – “Fluid Flow Design Methods”
.
Industrial & Engineering Chemistry
29
(
5
),
584
584
.
Genereaux
R. P.
1937b
Fluid-flow design methods
.
Industrial and Engineering Chemistry
29
(
4
),
385
388
.
Gray
H. F.
1940
Sewerage in ancient and mediaeval times
.
Sewage Works Journal
12
(
5
),
939
946
.
Haut
B.
Vivier
D.
2012
Water supply in the middle east during Roman and Byzantine periods
. In:
Evolution of Water Supply Through the Millennia
(
Angelakis
A. N.
Mays
L. W.
Koutsoyiannis
D.
Mamassis
N.
, eds).
IWA Publishing
,
London, UK
.
Hodge
A. T.
2002
Roman Aqueducts & Water Supply
. 2nd edn,
Duckworth
,
London, UK
.
Jeffcott
H. H.
1928
No. 4. The determination of the most economic size of pipe-line for water-power installations
. In:
The Scientific Proceedings of the Royal Dublin Society
.
The Royal Dublin Society
,
Dublin, Ireland
, pp.
37
48
.
Koutsoyiannis
D.
Angelakis
A. N.
2003
Hydrologic and hydraulic science and technology in Ancient Greece
. In:
The Encyclopedia of Water Science
(
Stewart
B. A.
Howell
T.
, eds).
Marcel Dekker
,
New York, USA
, pp.
415
417
.
Lansey
K. E.
2006
The evolution of optimizing water distribution system applications
. In:
8th Annual Water Distribution Systems Analysis Symposium. Cincinnati
,
OH, USA
,
27–30 August 2006
.
Lansey
K. E.
Mays
L. W.
1989
Optimization models for design of water distribution systems
. In:
Reliability Analysis of Water Distribution Systems
(
Mays,
L. R.
ed.).
ASCE
,
New York, NY
, pp.
37
84
.
Lischer
V. C.
1948
Determination of economical pipe diameters in distribution systems
.
Journal AWWA (American Water Works Association)
40
(
8
),
849
867
.
Maier
H. R.
Kapelan
Z.
Kasprzyk
J.
Kollat
J.
Matott
L. S.
Cunha
M. C.
Dandy
G. C.
Gibbs
M. S.
Keedwell
E.
Marchi
A.
Ostfeld
A.
Savic
D.
Solomatine
D. P.
Vrugt
J. A.
Zecchin
A. C.
Minsker
B. S.
Barbour
E. J.
Kuczera
G.
Pasha
F.
Castelletti
A.
Giuliani
M.
Reed
P. M.
2014
Evolutionary algorithms and other metaheuristics in water resources: current status, research challenges and future directions
.
Environmental Modelling & Software
62
,
271
299
.
Manning
R.
Griffith
J. P.
Pigot
T. F.
Vernon-Harcourt
L. F.
1891
On the flow of water in open channels and pipes
.
Transactions of the Institution of Civil Engineers of Ireland
20
,
161
207
.
Martin
D. W.
Peters
G.
1963
The application of Newton's method to network analysis by digital computer
.
Journal of the Institute of Water Engineers
17
(
2
),
115
129
.
Mays
L. W.
1989
Reliability Analysis of Water Distribution Systems
.
American Society of Civil Engineers (ASCE)
,
New York, USA
.
Mays
L. W.
2000
Water Distribution Systems Handbook
.
McGraw-Hill
,
New York, USA
.
Mays
L. W.
2013
A brief history of water filtration/sedimentation
.
Water Science and Technology: Water Supply
13
(
3
),
735
742
.
Mays
L. W.
Koutsoyiannis
D.
Angelakis
A. N.
2007
A brief history of urban water supply in antiquity
.
Water Science and Technology: Water Supply
7
(
1
),
1
12
.
Mays
L. W.
Sklivaniotis
M.
Angelakis
A. N.
2012
Water for human consumption through history
. In:
Evolution of Water Supply Through the Millennia
(
Angelakis
A. N.
Mays
L. W.
Koutsoyiannis
D.
Mamassis
N.
, eds).
IWA Publishing
,
London, UK
.
Moody
L. F.
Princeton
N. J.
1944
Friction factors for pipe flow
.
Transactions of the ASME
66
(
8
),
671
684
.
Needham
J.
1965
Science and Civilisation in China, Volume 4: Physics and Physical Technology, Part II: Mechanical Engineering
.
Cambridge University Press
,
London
.
Nevons, H.
1889
Address of the president
.
Journal of New England Water Works Association (JNEWWA)
4
(
1
),
3
5
.
Nicklow
J.
Reed
P.
Savic
D.
Dessalegne
T.
Harrell
L.
Chan-Hilton
A.
Karamouz
M.
Minsker
B.
Ostfeld
A.
Singh
A.
Zechman
E.
2010
State of the art for genetic algorithms and beyond in water resources planning and management
.
Journal of Water Resources Planning and Management, ASCE
136
(
4
),
412
432
.
Ormsbee
L. E.
2006
The history of water distribution network analysis: The computer age
. In:
8th Annual Water Distribution Systems Analysis Symposium
.
Cincinnati, Ohio, USA
,
27–30 August 2006
.
Ormsbee
L. E.
Lansey
K. E.
1994
Optimal control of water supply pumping system
.
Journal of Water Resources Planning and Management, ASCE
120
(
2
),
237
252
.
Ostfeld
A.
Shamir
U.
1993
Incorporating reliability in optimal design of water distribution networks - review and new concepts
.
Reliability Engineering & System Safety
42
(
1
),
5
11
.
Pollio
M. V.
Perrault
C.
Tardieu
E.
1859
Les Dix Livres d'Architecture de Vitruve
. A.
Morel et Cie
,
Paris, France
(
in French)
.
Pope
J. M.
Weir
M. H.
Rose
J. B.
2012
History of water and health
. In:
Evolution of Water Supply Through the Millennia
(
Angelakis
A. N.
Mays
L. W.
Koutsoyiannis
D.
Mamassis
N.
, eds).
IWA Publishing
,
London, UK
.
Ramalingam
D.
Lingireddy
S.
Ormsbee
L. E.
2004
History of water distribution network analysis: over 100 years of progress
. In:
Environmental and Water Resources History, ASCE 150th Anniversary Conference
(
Rogers
J. R.
Fredrich
A. J.
, eds).
ASCE
,
Washington, DC, USA
,
3–7 November 2002
, pp.
55
67
.
Rossman
L. A.
1993
EPANET Users Manual
.
Risk Reduction Engineering Laboratory, US Environmental Protection Agency (EPA)
,
Cincinnati, Ohio
(
June 1993)
.
Rouse
H.
Ince
S.
1963
Water transport section
. In:
Water Storage, Transport, and Distribution
(
Takahasi
Y.
, ed.).
Encyclopedia of Life Support Systems (www.eolss.net
).
Developed under the auspices of the UNESCO, EOLSS Publishers
,
Oxford, UK
.
Sanks
R. L.
2005
Water Transport section. In: Water Storage, Transport, and Distribution (Y. Takahasi, ed.). Encyclopedia of Life Support Systems (www.eolss.net). Developed under the auspices of the UNESCO, EOLSS Publishers, Oxford, UK
.
Sarchet
B. R.
Colburn
A. P.
1940
Economic pipe size in the transportation of viscous and nonviscous fluids
.
Industrial & Engineering Chemistry
32
(
9
),
1249
1252
.
Savic
D. A.
Banyard
J. K.
2011
Historical development of water distribution practice
. In:
Water Distribution Systems
(
Savic
D. A.
Banyard
J. K.
, eds).
ICE Publishing
,
London, UK
.
Savic
D. A.
Walters
G. A.
1997
Genetic algorithms for least-cost design of water distribution networks
.
Journal of Water Resources Planning and Management, ASCE
.
123
,
67
.
Shamir
U.
1974
Optimal design and operation of water distribution systems
.
Water Resources Research
.
10
(
1
),
27
36
.
Shamir
U.
1979
Optimization in water distribution systems engineering
.
Engineering Optimization
.
11
,
65
84
.
Simpson
A. R.
Dandy
G. C.
Murphy
L. J.
1994
Genetic algorithms compared to other techniques for pipe optimization
.
Journal of Water Resources Planning and Management, ASCE
.
120
(
4
),
423
443
.
Sklivaniotis
M.
Angelakis
A. N.
2006
Water for Human Consumption through the History.
In:1st IWA International Symposium on Water and Wastewater Technologies in Ancient Civilizations
,
Iraklion, Greece
,
28–30 October 2006
, pp.
659
666
.
Smiles
S.
1862
Lives of the Engineers
.
John Murray
,
London
.
Snow
J.
1849
On the pathology and mode of communication of cholera
.
London Medical Gazette or Journal of Practical Medicine
9
, p745–752 and
923–929
.
Snow
J.
1855
On the Mode of Communication of Cholera
.
John Churchill
,
London, UK
.
Spiess
O. V.
1887
Graphische lösung von hydraulischen aufgaben (Graphical Method for Hydraulic Problems)
.
Journal fur Gasbeleuchtung und Wasserversorgung
17
,
563
574
(in German)
.
Taha
H. A.
1992
Operations Research: An Introduction
.
Prentice Hall International Editions
,
Macmillan Publishing Company
,
Englewood Cliffs, NJ
.
Tavallaee
A.
1974
Inclusion of Pumps, Pressure Reducing Valves and Reservoirs in Pipe Networks Solved by Linear Theory Method
.
Master Thesis
.
Department of Civil and Environment Engineering, Utah State University
,
Logan, UT
.
Todini
E.
Pilati
S.
1988
A Gradient Method for the Solution of Looped Pipe Networks
.
Computer Applications in Water Supply: Vol. 1 (System analysis and simulation)
,
Wiley
,
London
, pp. 1–20.
True
A. O.
1937
Economical sizes for water force mains
.
Journal AWWA (American Water Works Association)
29
(
4
),
536
547
.
Tuttle
G. W.
1895
The economic velocity of transmission of water through pipes
.
Engineering Record
XXXII
(
15
),
258
258
.
Van Dijk
M.
Van Vuuren
S. J.
Van Zyl
J. E.
2008
Optimising water distribution systems using a weighted penalty in a genetic algorithm
.
Water SA
34
(
5
),
537
548
.
Viollet
P.-L.
2000
L'Hydraulique dans les Civilisations Anciennes: 5000 Ans d'Histoire
.
Presses de l'Ecole National des Ponts et Chaussees
,
Paris, France
(
in French)
.
Walski
T. M.
1985
State-of-the-Art Pipe Network Optimization. Computer Applications in Water Resources
.
ASCE
,
Buffalo
,
New York, June 1985
, 559–568 pp.
Walski
T. M.
2006
A history of water distribution
.
Journal AWWA (American Water Works Association)
98
(
3
),
110
121
.
Walski
T. M.
Chase
D.
Savic
D.
2006
Water distribution system analysis before digital computers
. In:
8th Annual Water Distribution Systems Analysis Symposium
.
Cincinnati, Ohio, USA
,
27–30 August 2006
.
Walters
G. A.
1992
A review of pipe network optimization techniques
. In:
Pipeline Systems, Springer
,
Dordrecht, The Netherlands
pp.
3
13
.
Wikipedia
2014
(accessed 7 August 2014).
Wilson
C.
1995
The Invisible World: Early Modern Philosophy and the Invention of the Microscope
.
Princeton University Press
,
Princeton, NJ
.
Wood
D. J.
1980
Computer Analysis of Flow in Pipe Networks Including Extended Period Simulations: User's Manual
.
Office of Continuing Education and Extension of the College of Engineering of the University of Kentucky
,
Lexington, KY
.
Wood
D. J.
Charles
C. O. A.
1972
Hydraulic network analysis using linear theory
.
Journal of the Hydraulics Division
98
(
HY7
),
1157
1170
.