Dairy Science & Technology · Dairy Engineering & Plant Operations · also Dairy Plant Design and Layout, Process Equipment Design, Project Planning
Planning a dairy plant from product mix to building: location, layout, equipment sizing, pressure vessels, costing and commissioning.
In the exams: ICAR AIEEA-PG group 17 Unit II (dairy plant layout and design, functional design, space requirement); ICAR AICE subject 38 Unit 5
Dairy plant design is the work of turning a decision to process milk into a building that receives, processes, packs and dispatches it safely and at a cost the business can carry. It starts with questions that are economic rather than mechanical: how much milk will arrive, in which season, at what composition, and which products will be made from it. Those answers fix the capacity of every line, the size of every silo and the floor area of every hall. The layout then arranges these spaces so that milk moves forward from raw to finished without crossing back, people and air move in the directions hygiene requires, and services such as steam, refrigeration, water and effluent reach every point of use. Finally the project is costed, built, installed, commissioned and validated before the first litre is sold.
The subject draws on several others. Hygienic design of individual equipment items and cleaning-in-place are taught in this track's cip-sanitation-hygienic-design; the construction and operation of separators, homogenisers, heat exchangers and fillers in dairy-processing-equipment; the food safety system that zoning supports in dairy-food-safety-systems; and the financial appraisal of a project in dairy-business-management and the agribusiness subject financial-management. General food plant design, materials and machine elements are covered in the agricultural engineering subjects food-quality-safety-plant-design and machine-design, and food safety management systems in the food technology subject food-safety-management-systems. This manual gives short reminders of that theory and spends its words on what is particular to a dairy plant: a perishable raw material arriving daily in a seasonal pattern, wet processing with heavy cleaning loads, products ranging from chilled milk to dry powder, and a regulatory expectation that the building itself prevents contamination.
Plant design is examined in the ICAR AIEEA-PG Dairy Technology paper (group 17, Unit II: dairy plant layout and design, functional design and space requirement), in the ICAR AICE-JRF/SRF Dairy Engineering paper (subject 38, Unit 5: process equipment and pressure vessel design), in GATE Agricultural Engineering for its dairy and food engineering calculations, and in bank and cooperative recruitment where officers appraise dairy plant projects. Questions range from recalling the sequence of a layout or the purpose of a holding tube to calculating silo volume, shell thickness, pump power or plant capacity from stated data. Codes and norms change and differ by country; where this manual uses a number it is either a standard engineering relation or a value stated as an example, and a real design must follow the current edition of the code that applies at the site.
Location fixes most of a plant's lifetime costs before any equipment is bought. A dairy plant depends on a daily inflow of a perishable raw material, a large and reliable water supply, an outlet for treated effluent, electric power and fuel for steam, and road access for both milk tankers and product distribution. Because milk is about 87 percent water and spoils quickly unless chilled, transporting raw milk long distances is costly, while many finished products, especially powder, butter and ghee, are concentrated and keep longer. This gives the classic location rule for dairying: plants that make liquid milk for a city are pulled toward the market, and plants that convert surplus milk into storable products are pulled toward the milk shed.
The general theory of facility location, including factor rating and the centre-of-gravity method, is covered in the agribusiness subject production-operations-management. In dairy practice the decision is usually made in two steps. First a region is chosen from the balance of milk supply, market demand and policy (for example the area a cooperative union serves, or a state's industrial incentives). Then individual sites within that region are compared on the specific factors below.
The site factors that matter most for a dairy plant can be grouped into six. Raw material: the size, density and seasonality of the surrounding milk shed, the distance and road quality to chilling centres and bulk milk coolers, and competition from other buyers. Water: a dairy plant uses several litres of water for every litre of milk processed, mainly for cleaning, cooling towers and boiler feed, so the quantity, quality and legal right to abstract water must be confirmed before purchase; the actual ratio varies widely with plant type and management. Effluent: dairy wastewater has a high organic load, so the site needs space for a treatment plant and a permitted discharge route, whether to a sewer, a river or land. Power and fuel: a reliable grid connection of adequate capacity, and fuel for boilers. Transport: all-weather roads for tankers and distribution vehicles, and for a large plant possibly rail or port access for powder exports. Land and surroundings: enough area for the plant, future expansion, effluent treatment, parking and truck manoeuvring; ground that bears heavy silos and does not flood; prevailing wind direction relative to sources of dust, odour or smoke; and distance from contamination sources such as waste dumps, tanneries or chemical works.
Regulatory and social factors complete the list: zoning and land-use permission, environmental clearance, distance from residential areas because of noise, odour and truck traffic, and the availability of skilled labour. Many countries require an environmental impact assessment for larger dairy plants; the threshold and process differ by jurisdiction.
Candidate sites are usually compared with a weighted factor rating. Each factor receives a weight reflecting its importance, each site is scored on each factor, and the weighted scores are summed. The method is simple, but its value lies in making the weights explicit so that a board or funding agency can see why one site was preferred. Costs that can be quantified, such as milk collection distance, land price and the cost of laying a water main or a power line, should also be converted into annual cost and compared directly; a site that is cheap to buy but needs a long pipeline or a new substation may cost more over the project life.
Two dairy-specific checks are often forgotten. The first is expansion: dairy plants commonly grow in stages, so the site plan should reserve land for additional silos, a second processing hall or a drying tower, and the first-phase layout should not block them. The second is the effluent route: an otherwise ideal site can be ruled out if there is nowhere to discharge treated water or spread sludge, and retrofitting effluent capacity later is expensive.
Capacity planning begins with the milk that will arrive, not with the products the marketing team would like to sell. The planner needs the expected daily intake in the flush (peak) season and the lean season, its average fat and solids-not-fat (SNF), and how intake is expected to grow over the project life. In many tropical and subtropical milk sheds the flush-season intake can be well above the lean-season intake; the ratio depends on breed, feeding, climate and calving pattern and must be taken from local data.
The product mix then allocates this milk. A typical approach is to meet fresh-milk and fresh-product demand first, because these earn the most and cannot be stored, and to convert the remainder into storable products such as butter, ghee, skim milk powder or whole milk powder. Because demand for liquid milk is fairly steady while intake swings with the season, the conversion plant takes up the difference: in the flush it runs at full capacity, and in the lean season it may run part-time or stop. This is why a dairy plant often has lines of very different utilisation, and why the conversion section is sized for the flush surplus rather than for the average.
Fat and SNF must be balanced across the whole mix. Standardising liquid milk to a lower fat content than the raw milk releases cream, which goes to butter or ghee; making a low-fat product releases more. Skimmed milk not needed for liquid products goes to powder. The planner therefore runs a fat and SNF balance for each season to check that the chosen mix uses all the milk without shortfall or unsellable surplus of either component.
Every line has a rated capacity, usually expressed in litres per hour for a continuous process such as pasteurisation or separation, or in batches per day for batch equipment such as cheese vats. The daily capacity of a continuous line is its hourly rate multiplied by the productive hours per day, which is the shift length minus time for start-up, product changeovers and cleaning. Cleaning is significant in dairy plants: a pasteuriser is typically cleaned at least once a day, and heat exchangers on high-temperature duties need more frequent cleaning as deposits build up. If a plant must process 200,000 litres in a day with 16 productive hours, its pasteuriser must handle at least 12,500 litres per hour, and the designer then adds a margin for growth and surges.
Capacity utilisation is actual throughput divided by rated capacity over the same period. A plant whose conversion section runs only in the flush season will show low annual utilisation even if it is full when it runs, and this must be built into the project's financial projections. The agribusiness subject production-operations-management treats capacity measures in general; the dairy-specific point is that seasonality is the main driver of under-utilisation, and that strategies such as flattening the intake curve with price incentives, adding products that use surplus milk, or making products for stock can raise it.
A plant is a chain of operations, and the slowest link limits the whole. Reception, storage, processing, filling and cold storage must be matched so that no stage starves or blocks another. Reception capacity must handle the peak tanker arrivals, which often bunch in the morning after collection routes; raw milk silos buffer the difference between reception and processing. Between pasteuriser and filler, a buffer or surge tank allows the filler to stop briefly without stopping the pasteuriser. Fillers are usually rated in packs per hour, which must be converted to litres per hour using the pack size and checked against the pasteuriser rate allowing for filler efficiency. Cold store capacity must hold at least the finished stock between production and dispatch, plus a safety margin for transport delays.
When the product mix changes with the season, the bottleneck can move. In the lean season the liquid milk line may be the constraint; in the flush the drier or butter line. Planners therefore check the balance for each season, not only for an average day.
A process flow diagram (PFD) shows each unit operation as a block or symbol, connected by lines representing the flows of milk, products, by-products, utilities and cleaning solutions, with the main conditions such as temperature, flow rate and composition marked on each stream. The planner usually starts with a simple block diagram for each product: for pasteurised milk, reception, cooling, raw milk storage, clarification or separation and standardisation, homogenisation, pasteurisation and cooling, pasteurised milk storage, filling and cold storage. The block diagram is then developed into a PFD with stream numbers and a mass balance table, and later into piping and instrumentation diagrams (P&IDs) that show every valve, instrument and control loop.
The PFD is the reference document for sizing equipment, estimating utilities and checking that every stream has a destination. Dairy PFDs must show recirculation and diversion: the flow diversion valve on a pasteuriser returns under-temperature milk to the balance tank, and these streams affect pump and tank sizing. They should also show cleaning circuits, because cleaning-in-place (CIP) circuits are designed together with the process piping; the design of CIP itself is covered in cip-sanitation-hygienic-design.
A mass balance applies the conservation of mass to a process: what enters equals what leaves plus what accumulates. In steady continuous processing nothing accumulates, so total mass in equals total mass out, and the same holds separately for each component such as fat or SNF, provided no chemical reaction creates or destroys it. Most dairy plant calculations are of this type.
For a separator, whole milk splits into cream and skim milk. With whole milk mass W at fat fraction fw, cream C at fat fc and skim S at fat fs, the balances are W = C + S and W·fw = C·fc + S·fs. Solving gives C = W(fw − fs)/(fc − fs). For example, 10,000 kg of milk at 4.0 percent fat separated into 40 percent cream and 0.05 percent skim gives C = 10,000 × (4.0 − 0.05)/(40 − 0.05), about 989 kg of cream and 9,011 kg of skim.
Standardisation by mixing two streams uses the same structure, often solved with the Pearson square for two components. To make standardised milk of a target fat from whole milk and skim milk, the proportions are inversely related to the differences between each stream's fat and the target. In-line standardisation in a modern plant does this continuously by remixing a controlled flow of cream into skim milk after the separator.
Mass balances also give yields. The yield of butter from cream depends on the fat in cream, the fat in butter and the fat lost in buttermilk; the yield of skim milk powder depends on the total solids of skim milk and the moisture of the powder. For a powder, the dry matter balance is skim mass × skim total solids = powder mass × (1 − powder moisture), so 10,000 kg of skim milk at 9.0 percent total solids makes about 900 ÷ 0.96, or 937.5 kg, of powder at 4 percent moisture, before allowing for losses.
An energy balance applies the same logic to heat. For heating or cooling a liquid without phase change, the heat duty is Q = m·cp·ΔT, where m is the mass flow, cp the specific heat and ΔT the temperature change. Milk's specific heat is close to that of water, a little lower; designers use a stated value, commonly about 3.9 kJ/kg·K for whole milk. Regeneration in a plate pasteuriser recovers much of this heat: incoming cold milk is warmed by outgoing hot milk. Regeneration efficiency is the heat recovered in regeneration divided by the heat that would be needed to raise the milk from its inlet temperature to the pasteurisation temperature, which for equal flows reduces to a ratio of temperature differences. If milk enters at 5 °C, leaves the regeneration section at 65 °C and is pasteurised at 75 °C, the efficiency is (65 − 5)/(75 − 5), about 86 percent.
Utility balances total the steam, refrigeration, electricity, water and compressed air required by every unit, for each season and shift pattern, to size the boiler, refrigeration plant, transformers, water treatment and compressors. These are taught in detail in this track's dairy-plant-utilities-energy and dairy-refrigeration-cold-chain; in plant design the key point is that utility peaks often coincide with cleaning, when hot detergent and rinse water are drawn at the same time as production continues on other lines.
Functional design decides what spaces a plant needs and how big each must be before deciding where they go. It starts from the process flow diagram and the list of equipment, then adds every activity that must happen in the building: receiving and testing milk, storing it, processing, filling, packing, storing finished goods, dispatch, cleaning, maintenance, laboratory work, storage of packaging and ingredients, chemical storage, changing and washing for staff, offices, and the services plant. Each activity becomes a functional area with stated requirements: floor area, clear height, floor loading, temperature and humidity, hygiene level, drainage, lighting and services.
The output is a schedule of accommodation, a table listing every room with its area and requirements, together with an adjacency or relationship chart showing which areas must be close, which may be close and which must be kept apart. In a dairy plant, the reception dock must be next to raw milk storage; the filling room next to pasteurised milk storage and to the cold store; the chemical store close to the CIP station but separated from product areas; and toilets and changing rooms close to production entrances but never opening directly into a production room.
Space is estimated in three ways that are used together. The first is equipment-based: the plan area of each machine plus clearances on every side for operation, cleaning and maintenance, plus space to withdraw parts such as heat exchanger plates or separator bowls. Manufacturers state the minimum clearances for their equipment, and hygienic design guidance asks for enough space behind and under equipment to clean the floor. The second is activity-based: areas such as the laboratory, crate washing or a packaging store are sized from the work done in them, for example the number of benches, the crate throughput or the days of packaging stock held. The third is norm-based: planners use area norms from experience, such as square metres per thousand litres of daily capacity for a given type of plant, to check the totals. Norms are useful for early estimates and for checking consultants' proposals, but they differ between countries, plant types and levels of automation and should be treated as a cross-check, not a design basis.
Cold stores and dry stores are sized from the stock they must hold. For a cold store, the number of crates or pallets is the daily dispatch multiplied by the days of stock held, and the floor area follows from the pallet or crate footprint, the stacking height and an allowance for aisles. Aisle allowance can be large in practice, often a third or more of the floor, depending on the handling equipment.
Clear height matters as much as floor area. Processing halls need height for tall equipment such as silos inside the building, separators with bowl removal, and overhead pipe racks; powder plants need very tall buildings for drying chambers and fluid beds. Floor loading must be calculated for full tanks and silos, which carry large concentrated loads, and for pallet racking in stores. Designers also allow for flexibility: wider spans with fewer internal columns allow lines to be rearranged, and service corridors or overhead pipe bridges allow new lines to be connected without breaking into hygienic areas. A common error in small plants is a building designed exactly around the first equipment list, which then cannot accept a second line, a larger pasteuriser or a change of product without demolition.
Layout arranges the functional areas on the site and within the building. Three principles dominate dairy layouts. The first is forward flow: raw milk enters at one end, products leave at the other, and nothing moves backwards from a cleaner to a dirtier area. The second is separation of raw and processed product, so that pasteurised milk and products cannot be recontaminated by raw milk, its containers, its handlers or its air. The third is short, simple product paths, because every metre of pipe must be cleaned and every transfer is a chance for contamination or product loss.
Common layout shapes are the straight-line flow, where reception and dispatch are at opposite ends of a long building; the U-shape, where reception and dispatch are on the same side of the building but separated, sharing yard space and roads; and the L-shape, which fits corner sites. The straight line gives the clearest separation of raw and finished product but needs a long site and two road accesses. The U-shape is compact and keeps services central, but care is needed to prevent raw milk tankers and finished product vehicles crossing in the yard. General layout types, including product and process layouts, are covered in production-operations-management; dairy plants are mostly product-type layouts with dedicated lines, with process-type grouping in areas such as a laboratory or a maintenance workshop.
The reception area receives milk from tankers or, in smaller plants, in cans. Tanker reception needs a covered bay with a drained, washable floor, a sampling point, flow metering or weighbridge, a de-aerator before the meter, a filter, a cooler if the milk is not cold enough, and a tanker cleaning station so that tankers leave clean. Can reception needs a raised dock, can tipping, weighing and sampling, and a can washer. The milk testing laboratory should be next to reception so that samples are tested before milk is accepted. Raw milk silos are usually located outside the main building or in a dedicated silo room, close to reception and connected to the processing hall by short pipelines; outdoor silos need insulation and a weatherproof cladding, with their outlets and valve clusters inside a hygienic enclosure.
The processing hall houses separators, standardisation units, homogenisers, heat treatment plant and buffer tanks. It is usually a wet area with heavy cleaning, and it should allow access to every machine for operation and maintenance without walking through the filling room. Pasteurised product goes through short lines to buffer tanks and then to filling machines in a separate filling room, which is kept to a higher hygiene standard than the processing hall because product there is exposed or nearly so. From filling, packed product moves by conveyor directly into the cold store, preferably without passing through any less clean area, and from the cold store to a refrigerated dispatch dock.
Product-specific sections are arranged alongside. A cheese plant adds vat rooms, pressing, brining and ripening rooms with controlled temperature and humidity. A butter section adds cream ripening tanks, a churn or continuous butter maker and a packing room kept cool. A powder plant separates the wet side (evaporators and concentrate tanks) from the dry side (drier, fluid bed, powder handling and packing), as the next chapter explains. Ingredients, packaging materials and chemicals each need their own store, with packaging de-boxed in a separate area so outer cartons, which carry dust and pests, do not enter production rooms.
Hygienic zoning divides a plant into areas with different hygiene requirements and controls the movement of people, materials, product and air between them. The aim is to protect product at the points where it is most exposed. Terms vary between standards and companies: some use low-, medium- and high-care zones; others use basic hygiene, medium hygiene and high hygiene, or non-production, production and high-risk areas. What they share is a gradient of control. Outer areas such as offices, workshops and raw material stores need basic good manufacturing practice. Raw milk reception and processing areas where product is enclosed in pipes and tanks need a higher standard. Areas where heat-treated product is exposed before final packing, such as the filling rooms for fresh products, cheese moulding and packing, or the powder packing room, need the highest controls.
Zoning is one of the prerequisite programmes that support a food safety system; the system itself, including how zoning is documented and audited, is covered in dairy-food-safety-systems. In plant design, zoning determines walls, doors, air handling, drainage and changing facilities, and these are very expensive to change after construction. It is therefore decided at the layout stage, drawn on the plan with colour coding, and agreed with the quality team before the building is designed.
People are a major route of contamination, so personnel flow is designed with the same care as product flow. Staff should enter production through changing rooms where outdoor clothing is removed, hands are washed and production clothing, hair covering and footwear are put on. High-hygiene areas often have a second barrier at the entrance: a step-over bench or a clearly marked line separating outside and inside footwear, hand washing and sanitising stations that operate without touching taps, and sometimes boot washers. Toilets must not open directly into production areas. Staff who work in raw milk reception should not walk into filling rooms; routes are arranged so they do not need to, and colour-coded clothing makes breaches visible.
Materials follow similar rules. Packaging enters high-hygiene areas through hatches or air locks after outer wrappings are removed. Waste leaves by a separate route. Equipment and tools are dedicated to a zone and colour coded where practical. Forklifts and pallets used outside are kept out of high-hygiene rooms; product and materials are transferred across the boundary onto clean, internal pallets or conveyors.
Air carries dust, aerosols and microorganisms, so the cleanest rooms are supplied with filtered air and kept at a higher pressure than surrounding rooms. Air then flows outwards from clean to less clean areas through doors and gaps, never inwards. A pressure cascade from the filling room to the processing hall to corridors and outside is a typical design. The grade of filtration and the number of air changes depend on the product and risk; extended-shelf-life filling and infant formula powder packing demand much more than a cheese store. Condensation is another hazard: cold surfaces in humid rooms drip, and drips from ceilings and pipework over open product are a recognised contamination route. Ventilation, insulation and the routing of cold pipes are designed to prevent it.
Powder plants have a further distinction between wet and dry zones. Water is the enemy of a dry zone: wet cleaning in powder areas creates moist pockets where bacteria such as Salmonella and Cronobacter can grow and then contaminate dry product. Dry zones are therefore cleaned by dry methods such as vacuuming wherever possible, floors are kept dry, and wet cleaning is controlled and followed by thorough drying. The dry zone is physically separated from wet processing, and personnel and materials entering it pass through controlled access. This principle is especially strict in plants making powders for infant formula.
A pressure vessel is a closed container designed to hold a fluid at a pressure substantially different from atmospheric. In a dairy plant they include steam boiler drums and steam headers, air receivers, refrigeration receivers and separators, jacketed process tanks heated with steam, UHT and retort vessels, evaporator calandrias and some pressurised storage tanks. A pressure vessel stores energy, and its failure can kill, so its design, fabrication, inspection and testing are governed by codes and, in most countries, by law.
The most widely used design code is the ASME Boiler and Pressure Vessel Code, Section VIII, of which Division 1 gives rules for the design of most unfired pressure vessels. In Europe the Pressure Equipment Directive sets essential safety requirements and the harmonised standard EN 13445 gives design rules for unfired pressure vessels. India's standard for unfired pressure vessels is IS 2825, and boilers are regulated separately under boiler legislation. Other countries have their own codes or adopt one of these. A design must follow the current edition of the code that applies at the site; the code fixes allowable stresses for each material, joint efficiencies, minimum thicknesses, inspection and testing. This chapter explains the principles and the commonly taught shell formula, not the full code.
When the wall thickness is small compared with the diameter, the stresses in a cylindrical shell under internal pressure P can be found by simple equilibrium. The hoop (circumferential) stress is σh = PD/(2t), and the longitudinal stress is σl = PD/(4t), where D is the diameter and t the thickness. The hoop stress is twice the longitudinal stress, so a cylinder under pressure tends to fail along a longitudinal seam, and longitudinal welds are the critical joints. For a sphere of the same diameter and thickness, the stress in every direction is PD/(4t), half the hoop stress in a cylinder, which is why spherical shapes and hemispherical heads need only about half the thickness of a cylinder of the same diameter.
Design codes modify these thin-wall relations to give the required thickness. The ASME Section VIII Division 1 rule for a cylindrical shell under internal pressure, governed by circumferential stress, is t = PR/(SE − 0.6P), where R is the inside radius, S the maximum allowable stress of the material at the design temperature and E the joint efficiency of the longitudinal weld. A corrosion allowance is then added to the calculated thickness, and the plate thickness actually ordered is the next standard size above the total. Joint efficiency depends on the type of weld and the extent of radiographic examination, and is taken from the code.
A cylindrical vessel is closed with heads. Flat heads are the simplest but are very weak under pressure and need great thickness, so they are used only for low pressures or small diameters. Hemispherical heads are the strongest shape and need the least thickness, but they are deep and costly to form. Semi-ellipsoidal heads with a 2:1 ratio of major to minor axis are a common compromise and, under the ASME rule, need about the same thickness as the shell they close. Torispherical heads, made of a spherical crown and a small knuckle radius, are shallower and cheaper and are widely used on process tanks, at the cost of a greater thickness than an ellipsoidal head for the same pressure. Conical heads and bottoms are used where drainage or solids discharge is needed, which in dairy vessels is often more important than pressure.
Openings for nozzles, manholes and instruments weaken the shell, so codes require reinforcement around openings above a certain size. Vertical vessels are supported on legs, a skirt or brackets (lugs), and horizontal vessels on saddles; supports must carry the weight of the vessel full of liquid, any wind or seismic load, and must not create stress concentrations that crack the shell. For hygienic dairy vessels, supports and legs must also allow cleaning of the floor underneath and avoid crevices, so hygienic design rules from cip-sanitation-hygienic-design apply alongside the pressure code.
Finished pressure vessels are tested, usually by a hydrostatic test at a pressure above the design pressure set by the code, and inspected periodically during service under national rules. Safety relief valves sized for the worst credible overpressure are mandatory.
Tank and silo size follows from the volume to be held and the time it must be held. A raw milk silo must hold the intake between deliveries and processing, often a day or more of reception in the flush. A buffer tank between pasteuriser and filler is sized to cover a filler stoppage of a stated number of minutes at the pasteuriser's flow rate. The working volume is then increased for headspace, which allows for foaming, agitation and level control; the total volume is working volume divided by the fraction usable. A tank holding 50,000 litres of working volume at 90 percent fill needs about 55,600 litres of total capacity.
The dimensions follow from geometry. For a vertical cylinder, V = πD²H/4. Silos are tall and slender to save floor space, with heights several times their diameter, while process tanks are more nearly square in proportion to allow good agitation. The bottom is sloped or conical for drainage. Silos have agitators to prevent cream separation, level and temperature indicators, a manhole, a vent with a filter, spray balls for CIP, and a sampling valve. The construction details are in dairy-processing-equipment.
Continuous equipment such as separators, homogenisers, pasteurisers and UHT plants is rated by hourly capacity. The required capacity is the volume to be processed divided by the productive running hours, with a margin for growth. Lines are often specified to match: a separator, homogeniser and pasteuriser on the same line are given the same nominal capacity so they can be run in series. If the homogeniser is used only for the standardised milk and not for all the product, its capacity follows the flow it actually treats.
In a pasteuriser, the holding tube must hold every particle of milk at the pasteurisation temperature for at least the required time. The holding time of the average particle is the tube volume divided by the volume flow rate. But flow in a tube is not uniform: the fastest particles at the centre travel faster than the mean. In fully turbulent flow, the maximum velocity is commonly taken as about 1.2 times the mean, and in laminar flow as twice the mean. The tube length is therefore calculated for the fastest particle: length = required time × maximum velocity. For example, for a flow of 10,000 litres per hour in a tube of 48 mm inside diameter and a minimum holding time of 15 seconds, the mean velocity is about 1.54 m/s, so in turbulent flow with a factor of 1.2 the length is about 27.6 m. Regulations in some countries specify how the holding time is measured and verified.
Pipe diameter is chosen from the flow rate and a target velocity. Too low a velocity leaves the pipe poorly cleaned during CIP and allows product to separate; too high wastes pumping energy and can damage fat globules or create foam. CIP needs a minimum flow velocity in the pipe, commonly cited as about 1.5 m/s, which often determines the pipe size or the CIP pump size.
Pump power is found from the flow and the total head. The hydraulic power is ρgQH, where ρ is density, g the acceleration due to gravity, Q the volume flow and H the total head; the shaft power is the hydraulic power divided by the pump efficiency, and the motor is chosen from the next standard size above. Centrifugal pumps are standard for milk and CIP liquids; positive displacement pumps, such as lobe pumps, are used for viscous products like cream, cultured products and cheese curd, and for metering.
Equipment selection weighs more than capacity. Hygienic design and certification, materials of construction (usually austenitic stainless steel such as type 304 or the more corrosion-resistant type 316 for chloride-containing products and CIP), energy consumption, cleanability, automation, spare parts and service support, and the total cost of ownership over the equipment life all count. Buying a cheaper machine that costs more to clean, repair or run is a common error.
The building must protect product, be easy to clean, resist the chemicals and temperatures used, and last. Surfaces in production areas are smooth, impervious, non-absorbent, light in colour so dirt is visible, and resistant to water, steam, detergents, acids and sanitisers. Walls are commonly finished with insulated sandwich panels, glazed tiles, or impervious resin coatings over masonry; joints are sealed, and junctions between walls and floors are coved so dirt cannot collect in a right-angled corner. Ceilings are smooth and sealed, with no exposed beams or ledges above open product; false ceilings must be accessible for inspection and cleaning. Windows in production areas are fixed or screened, with sloping sills. Doors are tight-fitting, of impervious material, and often self-closing or fast-acting between zones.
Structural steel and pipe racks above production are minimised or enclosed, because horizontal surfaces collect dust and condensation. Where they cannot be avoided, they should be accessible for cleaning. Lighting fittings are sealed and shatter-protected over product areas. Insulation and vapour barriers are critical in cold stores and cooled rooms to prevent condensation and ice.
Dairy floors take heavy wear: wheeled traffic, dropped equipment, hot water, thermal shock from hot cleaning followed by cold rinsing, and chemical attack from lactic acid in spilled milk, cleaning acids and alkalis. Suitable finishes include acid-resistant tiles laid on a resistant bedding, and seamless resin floors such as polyurethane cement systems that tolerate thermal shock. The choice depends on traffic, temperature and chemicals, and the floor must be laid on a sound, well-drained base.
Floors must slope to drains so water does not pond. Slopes in wet processing areas are typically between about 1 in 100 and 1 in 50, depending on the water load and the finish; the fall over a run is the run length divided by the gradient denominator, so a 6 m run at 1 in 60 falls 0.10 m. Too steep a slope is unsafe and makes equipment hard to level. Drains should be located away from equipment and walking routes where possible, be easy to clean, and carry water away quickly. Channel drains of hygienic design, with removable gratings and rounded internal corners, and point drains (gullies) with traps both have their uses; the hygienic design of drains and the problem of biofilms in drains are covered in cip-sanitation-hygienic-design. Drain lines must flow from cleaner to less clean areas, never the reverse, and the drainage from high-hygiene rooms should not pass under lower-hygiene areas where blockages could back up.
A dairy plant depends on services, and their layout is part of plant design. Steam is generated in a boiler house kept outside the production building, with distribution mains, condensate return and traps; steam for direct contact with product, as in steam infusion or injection, must be culinary quality. Refrigeration supplies chilled water or ice water for product cooling and direct or secondary refrigerant to cold stores; ammonia systems are common in large plants and require machinery rooms with ventilation, detection and safe separation from occupied areas. Water supply includes potable water for product contact and cleaning, softened water for boilers, and recovered water reused where permitted. Compressed air for instruments and for product contact must be dry and oil-free where it touches product. Electrical distribution includes transformers, standby generation for critical loads such as refrigeration and control systems, and hygienic, washdown-rated fittings in production areas.
Effluent from processing and cleaning is collected separately from storm water and sewage and taken to treatment, covered in dairy-effluent-environment. Services are routed in dedicated service corridors, above ceilings or on external pipe bridges where possible, so that maintenance staff can work on them without entering high-hygiene rooms. Utility sizing itself is covered in dairy-plant-utilities-energy and dairy-refrigeration-cold-chain.
The capital cost of a dairy plant includes land and site development, buildings and civil works, plant and machinery, utilities and services, effluent treatment, laboratory, vehicles, installation and commissioning, engineering and consultancy fees, preliminary and pre-operative expenses, contingencies, and the margin money for working capital. Estimates become more accurate as design proceeds: an order-of-magnitude estimate at the idea stage, a budget estimate from the preliminary design, and a definitive estimate from detailed design and supplier quotations.
At early stages, cost is often scaled from a known plant using a capacity exponent. The six-tenths rule states that cost varies with capacity raised to about the power 0.6: cost₂ = cost₁ × (capacity₂/capacity₁)^0.6. Doubling capacity therefore raises cost by about 2^0.6, or 1.52 times, which reflects economies of scale. The exponent differs between types of equipment and plant and is a rough tool; costs must also be updated for inflation with a cost index when the reference plant was built in an earlier year.
A feasibility study tests whether the project is worth doing. It examines the market (demand, prices, competition), the technical design (capacity, process, location, utilities), the financial case (costs, revenues, cash flow, returns), the economic and social case where public money or cooperative members are involved, and the environmental and regulatory requirements. The financial appraisal uses the methods taught in dairy-business-management and the agribusiness subject financial-management: payback period, net present value, internal rate of return, benefit–cost ratio, and for lenders the debt service coverage ratio. A plant project is also tested with sensitivity analysis on the variables that most affect dairy plants: milk procurement price, product prices, capacity utilisation and the seasonal intake pattern.
Break-even analysis shows the capacity utilisation at which revenue covers total cost. With annual fixed cost F, and contribution per unit (price minus variable cost) c, break-even output is F/c; dividing by the plant's annual capacity gives the break-even utilisation. Because seasonal intake already limits utilisation, a dairy project whose break-even utilisation is close to its achievable utilisation is risky.
A detailed project report (DPR) is the document submitted to a board, a funding agency or a bank to obtain approval and finance. Its usual contents are a summary; the promoters and their background; the market analysis; the milk procurement plan with the milk shed, seasonal intake and price assumptions; the product mix and capacity; the process description and flow diagrams; the plant layout and building plan; the list of plant and machinery with specifications and quotations; utilities and effluent treatment; manpower; the project implementation schedule; the cost of the project and means of finance; the profitability projections, cash flow, balance sheet and repayment schedule; the financial appraisal and sensitivity analysis; environmental and regulatory clearances; and a risk assessment. Lenders such as development banks and cooperative funding agencies publish model project formats and check the assumptions against their own norms. A good DPR makes every assumption explicit and traceable so that it can be checked and updated.
Once a project is approved and equipment ordered, execution follows a schedule, commonly planned with bar charts and network methods such as the critical path method, which identifies the chain of activities that determines the earliest completion date. Civil works, building, utilities, equipment delivery, installation, piping, electrical and control work must be sequenced so that, for example, large silos and tanks are placed before walls are closed, and service connections are ready when equipment arrives.
Installation of dairy equipment must preserve hygienic design. Pipes are laid with slopes for drainage and without dead legs, welded by qualified welders using orbital welding where possible, with internal welds inspected, for example by borescope, because a poor internal weld is a crevice that cannot be cleaned. Equipment is levelled so that it drains. Insulation, supports and penetrations through walls are sealed. Every installation step is documented so that the as-built drawings match what was installed.
Commissioning brings the installed plant into operation and proves that it meets its design. It is often structured in stages. The factory acceptance test (FAT) is carried out at the supplier's works before shipment to check that equipment is built and functions as specified. The site acceptance test (SAT) checks the same after installation at site. Commissioning then proceeds from water runs, in which the plant is operated with water to check flows, temperatures, controls and leak-tightness, to product trials with milk, and finally to performance tests in which the plant must demonstrate its rated capacity, product quality and utility consumption over an agreed period.
In regulated industries a parallel structure of qualification is used. Installation qualification (IQ) documents that the equipment is installed as designed, with correct materials, connections and calibrated instruments. Operational qualification (OQ) shows that it operates as intended across its operating range, including alarms and interlocks. Performance qualification (PQ) shows that it consistently produces product meeting requirements under normal operation. The terms come from pharmaceutical practice and are increasingly used for high-risk dairy lines such as infant formula and aseptic processing.
Validation is documented evidence that a process consistently achieves its intended result. In a dairy plant the most important example is the pasteuriser, which is a critical control point in the food safety system. Validation of a pasteuriser includes checking that the holding time of the fastest particle meets the requirement, often by a salt or dye conductivity test that measures the time for a tracer to pass through the holding tube at the maximum flow rate; that the temperature recording and indicating thermometers are accurate and calibrated; that the flow diversion valve diverts milk at the set temperature and that the diversion is recorded; that the regeneration section is protected against leakage from raw to pasteurised milk by keeping the pasteurised side at a higher pressure; and that the flow rate cannot exceed the validated maximum. The specific tests and their frequency are laid down by regulators in many countries, for example in the US Pasteurized Milk Ordinance, and must be followed where they apply.
Other processes requiring validation include CIP, where cleaning effectiveness is verified as described in cip-sanitation-hygienic-design; UHT and aseptic filling, where sterility is demonstrated; and the cleaning and drying of powder plants. Validation is repeated after significant changes to equipment or process, and monitoring and verification then confirm in routine operation that the validated conditions are maintained.
A student uses this subject to connect the separate courses on processing, equipment, utilities and hygiene into one plant. Typical tasks are drawing a block diagram and layout for a pasteurised milk plant of stated capacity, running fat and SNF balances for a product mix, sizing a silo and a holding tube, and calculating the thickness of a jacketed tank shell. In-plant training is the time to walk the product, personnel and air routes of a working plant and compare them with the principles here; most final-year design projects and many viva questions ask exactly this.
Five years into a career, an engineer or production manager uses plant design when adding a line, relocating equipment, or fixing a recurring hygiene problem caused by layout. Decisions include whether a new filler fits the existing buffer tank and cold store, where to route a new pipe without creating a dead leg or crossing a zone boundary, how to change floors or drains in a running plant, and what commissioning and validation records an auditor will expect after a change. The practical skill is to see a layout problem behind a quality complaint, such as condensation over a filler or a drain backing up from a raw area.
A plant head or projects director with twenty-five years of experience decides on capacity expansions, new plants and major refurbishments. They set the design basis, choose between expanding an old site and building new, approve the layout and zoning concept, manage consultants and contractors, and defend the detailed project report before a board or lender. The key judgements are capacity against seasonal milk supply, phasing to keep utilisation high, hygienic standards appropriate to future products such as infant formula or aseptic milk, and lifetime cost rather than first cost. They also carry legal responsibility for pressure equipment, ammonia systems and worker safety.
A researcher or professor with long industry experience works on the methods behind design: optimisation of plant layout and scheduling under seasonal supply, life-cycle assessment and energy integration of dairy plants, hygienic design research on surfaces, drains and air handling, validation methods for new heat treatments, and the economics of scale and regional plant networks. They set examination papers and supervise design projects, review consultants' reports for public agencies, and contribute to standards committees and guidance on hygienic design and dairy plant engineering.
This manual gives the principles of dairy plant design and the commonly taught formulas. It does not reproduce the full pressure vessel codes (allowable stresses, joint efficiency tables, head and nozzle rules), national building and fire codes, or the detailed tests and frequencies that regulators set for pasteuriser validation; designers must use the current edition of the applicable code and regulation. Area norms per unit of capacity are not given because published norms differ widely by country, product and automation. Detailed utility sizing, refrigeration and effluent treatment design are left to their own subjects. Dynamic simulation, plant layout optimisation software and seismic design of tall silos are not covered.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions. Design values such as water ratios, area allowances, gradients and code formulas are stated in each question for the calculation; real designs must use local data and the current edition of the applicable code.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.