Agricultural Engineering · Post-Harvest, Food & Process Engineering · also Food Quality and Safety, Food Plant and Equipment Design, Process Plant Design
Measuring and assuring food quality, the food-safety systems regulators expect, and laying out a processing plant.
In the exams: ICAR AICE subject 41 Units 8 and 10; ICAR AIEEA-PG Unit VI (plant layout, feasibility reports); NCEES PE food safety
This subject joins three tasks that a food or agro-processing engineer does together. The first is quality: defining the attributes that make a product acceptable, measuring them, and using statistics to keep a process in control and to accept or reject lots. The second is safety: the hygiene practices, HACCP plans, preventive controls and food laws that regulators and buyers in every country now expect. The third is design: choosing where to build a plant, laying it out so that it is hygienic and efficient, drawing its flow sheets, sizing its equipment and utilities, judging whether it will pay, and dealing with its wastes and by-products.
The unit operations themselves, such as heat exchange, thermal processing, evaporation, drying, membranes and clean-in-place, are taught in the dairy and food engineering subject and are referred to here rather than repeated. This manual works at the level of the whole plant and the systems that govern it.
The subject is examined differently around the world. In India, ICAR's AIEEA-PG (Unit VI names plant layout and feasibility reports) and the AICE-JRF/SRF paper in Processing and Food Engineering ask both definitions and short calculations. In the United States, the PE Agricultural and Biological Engineering exam lists food safety, HACCP and HARPC under Processing Systems and food facilities under Facilities and Structures, and expects design calculations. Licensing exams in the Philippines, Bangladesh, Nepal and Canada include food processing and quality at undergraduate level. Laws differ by country; where this manual cites Indian, US, EU or Codex rules, check the current text before relying on it.
Food quality is the set of characteristics that decide whether a buyer accepts a product and whether it meets its specification. Textbooks group these attributes in three families. Sensory attributes are what the consumer perceives: appearance (colour, size, shape, gloss, freedom from defects), texture or kinaesthetic properties (firmness, crispness, viscosity, chewiness) and flavour (taste and aroma). Hidden attributes cannot be judged by the senses: nutritional value, the presence of toxins, pathogens or residues, and keeping quality. Quantitative attributes concern yield and the proportion of usable material, such as juice yield from fruit or milling yield from grain. Safety is not a grade of quality; a product that is unsafe is unacceptable however good it looks, which is why food-safety systems (later chapters) sit beside, not inside, ordinary quality control.
A specification turns attributes into numbers with limits: moisture not more than a stated percentage, net content not below a declared quantity, colour within a range, a microbiological criterion. The engineer's job is to choose a measurement for each attribute that is accurate (close to the true value), precise (repeatable), fast enough for the decision it supports, and affordable. An at-line refractometer reading every hour is often more useful for control than a laboratory method that returns results after the batch has shipped. Measurements also have to be traceable: instruments are calibrated against reference standards, methods follow a recognised standard method where one exists (official methods published by bodies such as AOAC International or ISO), and results are recorded so that a lot can be traced.
Colour is measured with a colorimeter or spectrophotometer in the CIE L*a*b* space. L* is lightness from 0 (black) to 100 (white); a* runs from green (negative) to red (positive); b* from blue (negative) to yellow (positive). From these, chroma (saturation) is the square root of a*² + b*², and hue angle is the arctangent of b*/a*. The total colour difference between a sample and a reference, ΔE*ab, is the square root of the sum of the squared differences in L*, a* and b*. Browning of a fruit purée, for example, shows as a fall in L* and a rise in a*.
Texture is measured with a universal testing machine. Texture profile analysis (TPA) compresses a sample twice, imitating two bites. Hardness is the peak force of the first compression; cohesiveness is the ratio of the work done in the second compression to that in the first; springiness is how far the sample recovers between bites; adhesiveness is the negative work as the probe pulls away. Gumminess is hardness × cohesiveness and chewiness is gumminess × springiness. Other instruments measure single properties: penetrometers for fruit firmness, shear cells for meat tenderness, rheometers and viscometers for fluids.
Composition and state are measured by moisture determination (oven drying is the reference for many products; results are expressed on a wet basis, water divided by total mass, or dry basis, water divided by dry matter), water activity (aw, the ratio of the vapour pressure of water in the food to that of pure water at the same temperature, equal to the equilibrium relative humidity divided by 100), total soluble solids by refractometer in °Brix, titratable acidity, pH, fat and protein. Near-infrared spectroscopy allows rapid, non-destructive estimates once it is calibrated against reference methods for that product.
Instruments cannot fully replace people, because the consumer is the final judge. Sensory tests fall in three classes. Discrimination tests ask whether a difference exists. In a triangle test each panellist receives three coded samples, two alike and one different, and must pick the odd one; the chance of a correct guess is one in three. In a duo-trio test a reference is followed by two coded samples, one matching it; the chance probability is one in two. Paired-comparison tests ask which of two samples has more of a stated attribute. The number of correct answers needed for significance is read from binomial tables for the chosen risk level.
Descriptive tests use a trained panel to rate the intensity of defined attributes on a line scale; quantitative descriptive analysis is the classic method. Affective (hedonic) tests use untrained consumers to rate liking, most often on the nine-point hedonic scale from 'dislike extremely' to 'like extremely'. Good practice applies to all of them: booths with controlled lighting, random three-digit codes, randomised serving order, palate cleansers, and enough panellists for the statistical test. Trained panels should never be used to measure consumer liking, and consumer panels cannot give reliable descriptive profiles.
Every process varies. Walter Shewhart distinguished common-cause variation, the steady background scatter built into the process, from special-cause (assignable) variation, which comes from something identifiable such as a worn filler valve or a new raw-material lot. A control chart plots a statistic from successive samples against a centre line and control limits placed three standard errors either side of it. A point outside the limits, or a non-random pattern inside them, signals a special cause worth investigating. Control limits describe what the process does; specification limits describe what the customer wants. The two must not be confused, and specification limits are never drawn on an X̄ chart of sample means.
For measured variables, the X̄ and R charts are used together. With k subgroups of size n, the grand mean X̿ and the average range R̄ give: X̄ chart limits X̿ ± A2·R̄; R chart upper limit D4·R̄ and lower limit D3·R̄. The constants depend on n; for n = 5, A2 = 0.577, D3 = 0 and D4 = 2.114; for n = 4, A2 = 0.729 and D4 = 2.282. The X̄ and s chart replaces the range by the standard deviation and is preferred for larger subgroups. Pattern rules add sensitivity: besides points beyond three-sigma limits, many plants act on a run of consecutive points on one side of the centre line. The length of the run that triggers action differs between rule sets (seven, eight or nine points are all in use), so a plant states which rule it applies.
For attributes (counts of defectives or defects), the p chart plots the fraction defective with limits p̄ ± 3√(p̄(1 − p̄)/n); the np chart plots the number defective for constant n; the c chart plots defects per unit with limits c̄ ± 3√c̄; the u chart plots defects per unit when the inspected quantity varies. A negative lower limit is set to zero.
A process can be in statistical control and still make product outside specification. Capability indices compare the spread of the process with the width of the specification. Cp = (USL − LSL)/6σ measures potential capability if the process were centred. Cpk = minimum of (USL − μ)/3σ and (μ − LSL)/3σ also accounts for off-centre running. A Cpk of 1.0 means the nearer specification limit is exactly three standard deviations from the mean. Customers and industries set their own minimum indices, often 1.33 or higher for critical characteristics; there is no single universal requirement. Capability should be calculated only for a process that is in control, from an estimate of σ obtained from the control chart or from enough individual results.
In filling operations capability links directly to law. Most countries regulate net quantity on packages, either by a minimum-quantity system (no pack may be below the declared amount) or by an average-quantity system in which the batch average must not be below the declared quantity and only a small proportion may fall below set tolerances. The fill target is then chosen from the process standard deviation, so a more precise filler lets the plant reduce overfill, which is a direct material saving.
Acceptance sampling decides whether to accept or reject a lot from a sample. In a single sampling plan for attributes (n, c), n items are inspected and the lot is accepted if the number of defectives is c or fewer. Double and multiple plans take a second sample when the first is inconclusive; sequential plans decide item by item. The operating characteristic (OC) curve shows the probability of acceptance (Pa) against the true fraction defective. For a large lot, Pa is the binomial probability of c or fewer defectives in n; the Poisson distribution with mean np is a common approximation. The producer's risk (α) is the probability of rejecting a lot at the acceptable quality level (AQL); the consumer's risk (β) is the probability of accepting a lot at the rejectable or limiting quality (LTPD or LQ). Tables such as ISO 2859-1 (attributes) and ISO 3951 (variables) index plans by lot size, inspection level and AQL. Sampling never guarantees a lot is defect-free; it controls the risks.
Microbiological criteria use special plans set out by the International Commission on Microbiological Specifications for Foods (ICMSF) and adopted in Codex principles. In a two-class plan (n, c, m), n sample units are tested and the lot is rejected if more than c units exceed the limit m; for pathogens such as Salmonella, c is commonly zero and m means 'absent in the analytical unit'. In a three-class plan (n, c, m, M), a result above M in any unit rejects the lot, and the lot is also rejected if more than c units fall between m and M. The values of n, c, m and M are set product by product and differ between countries' regulations.
HACCP grew out of the United States space programme in the 1960s, when a food company, NASA and the US Army laboratories needed near-certainty that astronauts' food was safe, and found that end-product testing could not give it. Its idea is to control hazards at the points in the process where they can be prevented, eliminated or reduced to acceptable levels, and to prove that control with records. The Codex Alimentarius text General Principles of Food Hygiene (CXC 1-1969, revised in 2020) is the international reference; it places HACCP on top of good hygiene practices, which must be working before a HACCP plan is built.
Codex lays out twelve steps. The first five are preliminary: assemble a multidisciplinary HACCP team; describe the product (composition, physical and chemical characteristics such as aw and pH, packaging, shelf life, storage and distribution); identify the intended use and users, including vulnerable groups; construct a flow diagram of every step; and confirm the flow diagram on site, during operation, at all hours the plant runs. Steps six to twelve apply the seven principles.
Principle 1 is to conduct a hazard analysis: list every biological, chemical (including allergens) and physical hazard reasonably likely at each step, assess its likelihood and severity, and identify control measures. Principle 2 is to determine the critical control points (CCPs), the steps at which control can be applied and is essential to prevent or eliminate a hazard or reduce it to an acceptable level. A decision tree helps but is not compulsory; the team's judgement decides. Principle 3 is to establish validated critical limits, the measurable criteria that separate acceptable from unacceptable, such as a minimum time and temperature, a maximum pH or a metal-detector reject of a stated test-piece size. Principle 4 is to establish a system to monitor control of each CCP, preferably continuous or frequent and fast (temperature, time, pH, flow), not slow microbiological testing. Principle 5 is to establish corrective actions for a deviation: restore control and decide what happens to the product made while control was lost. Principle 6 is to validate the HACCP plan and establish verification procedures, such as calibration, record review, targeted sampling and internal audit, to confirm that the system works as intended. Principle 7 is to establish documentation and record keeping.
Some terms need precision. Validation asks, before or at implementation, whether the control measures can control the hazard; verification asks, during operation, whether the plan is being followed and working. A critical limit is the regulatory or safety line; an operating limit is a tighter internal target that lets operators adjust before the critical limit is crossed. A CCP is not every important step: a pasteuriser's holding time and temperature is a classic CCP, while general cleaning is controlled by prerequisite programmes.
A working plan is short, specific and owned by the people on the line. Common failures are too many CCPs (which dilutes attention), critical limits copied from literature without validation for the actual equipment, monitoring records filled in after the fact, and plans that are not reviewed when a recipe, supplier or machine changes. A plan must be reviewed whenever anything changes that could affect a hazard, and at a set interval in any case.
Many countries make HACCP-based systems compulsory. In the European Union, food business operators other than primary producers must implement procedures based on HACCP principles under the hygiene regulation (EC) No 852/2004. In the United States, mandatory HACCP applies to meat and poultry under the US Department of Agriculture and to seafood and juice under the Food and Drug Administration; the juice rule requires processing that achieves a 5-log reduction of the most resistant pathogen of concern. Most other food under FDA now falls under the risk-based preventive controls of FSMA (chapter 5). In India, food business operators are required to follow the hygiene and sanitary practices of the licensing regulations, and FSSAI has required documented food-safety management systems for licensed businesses. Requirements change, so the current national text is the authority.
Good manufacturing practice (GMP), also called good hygiene practice (GHP), is the set of basic conditions and activities needed to keep a hygienic environment for producing safe food. In HACCP language these are prerequisite programmes (PRPs). They are not specific to one hazard at one step; they apply across the plant. Codex, ISO/TS 22002-1 and national regulations (Schedule 4 of India's licensing and registration regulations, Part 117 subpart B of the US rules, Annex II of EU Regulation 852/2004) list similar elements: the design and construction of buildings and equipment; water supply of potable quality; cleaning and sanitation; pest control; waste management; personal hygiene, health and training of staff; control of incoming materials and suppliers; storage and transport; allergen management; calibration and maintenance; product information, traceability and recall.
PRPs are what make a HACCP plan small. If pest control, glass-and-brittle-plastic control and hand hygiene are reliable, the team does not need CCPs for them. When a PRP fails, the hazard analysis has to be revisited.
Cleaning removes soil (food residues, fat, protein, mineral scale, biofilm); sanitation (disinfection) reduces the remaining microorganisms to a safe level. Sanitising a dirty surface does little, because soil shields microbes and inactivates many sanitisers. The standard sequence is dry clean-up, pre-rinse, detergent wash, rinse, sanitise and, where the sanitiser requires it, a final potable-water rinse. Four factors act together, often drawn as Sinner's circle: chemical action, temperature, mechanical action and time. Less of one must be made up by more of another.
Alkaline detergents remove fat and protein; acid detergents remove mineral scale such as milkstone and beerstone. Common sanitisers include chlorine compounds, quaternary ammonium compounds, peracetic acid, iodophors and hot water. Their effect depends on concentration, contact time, temperature, pH and water hardness, so each is used at the concentration and contact time stated by its supplier and permitted by national rules. Effectiveness is verified by visual inspection, ATP bioluminescence swabs (a rapid indicator of organic residue, not a count of microbes) and microbiological swabs. Clean-in-place for closed lines is covered in the dairy and food engineering subject; here the point is that cleaning is a validated process with records, not a chore.
People carry microbes. GMP requires medical fitness and exclusion of ill workers, hand washing at entry and after any contamination, protective clothing that stays in the production area, hair and beard covering, no jewellery, and controlled eating and smoking areas. Training must be recorded and refreshed.
Pest control relies first on exclusion (sealed openings, self-closing doors, screens, no harbourage), then on monitoring with mapped traps and bait stations, and only then on treatment by trained operators. Toxic baits are not used inside open-product areas.
Water used in contact with food or food-contact surfaces must meet national drinking-water standards, which differ by country; the World Health Organization guidelines for drinking-water quality are the common reference. Non-potable water for fire fighting or steam raising that does not touch food runs in separate, identified pipes with no cross-connection.
Plants are divided into hygiene zones. Raw-material handling is a low-risk area; areas where cooked or ready-to-eat product is exposed after the kill step are high-risk or high-care areas. Entry to high-care is through a changing area with a barrier, hand washing and dedicated clothing; air in high-care is filtered and kept at a slightly positive pressure so that air flows outward to less clean areas.
The US Food Safety Modernization Act (FSMA), signed in 2011, moved the Food and Drug Administration from responding to outbreaks to preventing them. It is implemented through a set of rules. The Preventive Controls for Human Food rule (21 CFR Part 117) updates current good manufacturing practice and requires most registered food facilities to have a written food safety plan based on hazard analysis and risk-based preventive controls, the approach widely called HARPC. A companion rule covers animal food. Other FSMA rules deal with produce safety on farms, the Foreign Supplier Verification Program for importers, sanitary transportation, mitigation of intentional adulteration (food defence) and additional traceability records for listed high-risk foods.
FSMA matters outside the United States because any facility that exports food to the US must register with FDA and meet these rules, and importers must verify that their foreign suppliers do. Engineers in Asia, Africa and South America working for exporters therefore apply HARPC directly.
A food safety plan under Part 117 contains a written hazard analysis, preventive controls, a supply-chain programme where a hazard is controlled by a supplier, a recall plan, and procedures for monitoring, corrective actions and verification. It must be prepared or overseen by a preventive controls qualified individual (PCQI), a person trained through a recognised curriculum or qualified by job experience.
The hazard analysis looks wider than classic HACCP. It must consider biological, chemical (including radiological) and physical hazards, and hazards that may be intentionally introduced for economic gain, such as adulteration of an ingredient to increase its weight or apparent quality. Preventive controls are of several kinds: process controls (which include, but are not limited to, CCPs), food allergen controls (preventing cross-contact and ensuring correct labels), sanitation controls (for environmental pathogens in ready-to-eat products exposed to the environment), supply-chain controls, and other controls. Not every preventive control needs a critical limit; the monitoring, corrective action and verification applied to each are scaled to its role in food safety. Process controls must be validated. Environmental monitoring and finished-product testing are named verification activities where appropriate. The plan must be reanalysed at least once every three years, and sooner when something changes.
The practical difference, then, is scope and flexibility: HACCP concentrates control at CCPs, whereas HARPC asks the facility to identify every preventive control needed and to manage each in proportion to the risk.
ISO 22000 specifies requirements for a food safety management system for any organisation in the food chain. Its current edition (2018) follows the common high-level structure of ISO management-system standards, so it can be integrated with ISO 9001 (quality) and ISO 14001 (environment). It combines interactive communication along the chain, system management, prerequisite programmes and HACCP principles. It uses two plan-do-check-act cycles, one for the management system and one for the operational hazard-control processes. It adds a category between PRPs and CCPs: the operational prerequisite programme (OPRP), a control measure applied to prevent or reduce a significant hazard where the measure is assessed by action criteria rather than a critical limit.
Retailers rely on third-party certification. The Global Food Safety Initiative (GFSI) benchmarks certification schemes against its requirements, and buyers commonly accept any GFSI-benchmarked scheme. Some schemes are built on ISO 22000 plus sector PRPs (ISO/TS 22002 series); others are independent standards. Certification is voluntary in law but often compulsory in trade. A certificate shows that a system was audited; it does not guarantee that every lot is safe.
The Codex Alimentarius Commission was created in 1963 by the Food and Agriculture Organization and the World Health Organization to develop international food standards, guidelines and codes of practice that protect consumer health and ensure fair practices in food trade. Its texts include commodity standards; general standards that apply across foods, such as the General Standard for the Labelling of Prepackaged Foods (CXS 1-1985), the General Standard for Food Additives (CXS 192-1995) and the General Standard for Contaminants and Toxins in Food and Feed (CXS 193-1995); maximum residue limits for pesticides and veterinary drugs; and codes of hygienic practice, including the General Principles of Food Hygiene.
Codex standards are voluntary, but they gained weight when the World Trade Organization's Agreement on the Application of Sanitary and Phytosanitary Measures (SPS Agreement, 1995) named Codex as the reference for food safety. A country that adopts a measure based on Codex is presumed to comply; a country that sets a stricter measure must justify it with a scientific risk assessment. Risk analysis in Codex has three parts: risk assessment (scientific, by expert bodies such as JECFA for additives and contaminants and JMPR for pesticide residues), risk management (policy choices by Codex committees and governments) and risk communication.
India's Food Safety and Standards Act, 2006 consolidated earlier food laws, including the Prevention of Food Adulteration Act, 1954 and several product orders (fruit products, milk and milk products, meat products, edible oils), into one law. It established the Food Safety and Standards Authority of India (FSSAI), which frames standards and regulations; enforcement is shared with state food safety commissioners and designated officers. Key regulations cover licensing and registration of food businesses, food product standards and food additives, contaminants, toxins and residues, packaging, labelling and display, food recall, import and laboratory sampling. Small (petty) food businesses register; larger ones need a state or central licence, with thresholds set by turnover, capacity and activity that FSSAI revises from time to time. The licensing regulations carry schedules on general hygienic and sanitary practices that act as India's legal GMP. Offences range from sub-standard and misbranded food to unsafe food, with penalties scaled to the harm.
In the United States, the Federal Food, Drug, and Cosmetic Act (1938) is the core law; FDA regulates most foods, while the US Department of Agriculture's Food Safety and Inspection Service regulates meat, poultry and processed egg products. Title 21 of the Code of Federal Regulations holds FDA's food rules: Part 117 (CGMP and preventive controls), Part 113 (thermally processed low-acid foods in hermetically sealed containers), Part 114 (acidified foods), Part 120 (juice HACCP), Part 123 (seafood HACCP) and Part 101 (labelling). Food additives require approval unless generally recognised as safe.
In the European Union, Regulation (EC) No 178/2002 (the General Food Law) lays down principles of risk analysis and the precautionary principle, places primary responsibility on food business operators, requires traceability one step back and one step forward, and created the European Food Safety Authority for risk assessment. The hygiene package, led by Regulation (EC) No 852/2004, sets general hygiene rules and the HACCP-based requirement, with specific rules for food of animal origin.
Every jurisdiction differs in details: permitted additives and limits, labelling formats, licensing thresholds, microbiological criteria. The engineer's rule is to identify every market a product will reach and check the current regulation of each, because a product legal in one country can be refused at the border of another.
Hygienic design means building equipment so that it does not contaminate product and can be cleaned easily and verifiably to a microbiologically safe state. A machine that is hard to clean will not be cleaned well, whatever the procedures say. The main bodies of guidance are the European Hygienic Engineering and Design Group (EHEDG), whose guideline documents are widely used worldwide; 3-A Sanitary Standards in the United States, historically for dairy and now wider; ISO 14159 (hygiene requirements for the design of machinery) and EN 1672-2 (food processing machinery, hygiene requirements). National regulations repeat the same principle in general words: food-contact surfaces must be smooth, non-absorbent, corrosion-resistant, non-toxic and easy to clean and disinfect.
Equipment is considered in zones. The product (food-contact) zone includes every surface that product, or anything that drips back into it, touches. The splash zone is reached by splashes that do not return to product. The non-product zone is the rest. Requirements are strictest in the product zone.
Austenitic stainless steels are the default for product-contact surfaces. AISI 304 is common; AISI 316, with added molybdenum, resists chlorides and acids better and is chosen for salty, acidic products and aggressive cleaning; low-carbon 'L' grades (304L, 316L) reduce sensitisation at welds. Plastics and elastomers in contact with food must be of food-contact grade under the relevant national rules and compatible with product fat, temperature and cleaning chemicals; elastomer seals are wear parts with replacement schedules. Wood, porous materials, mild steel, copper alloys exposed to acidic product, and painted product-contact surfaces are avoided.
Surface roughness is specified as Ra, the arithmetic mean deviation of the profile. EHEDG and 3-A guidance use Ra not exceeding 0.8 µm for product-contact surfaces, with exceptions allowed where cleanability is shown to be equivalent. Rougher surfaces hold soil and bacteria in the valleys. Finish quality also depends on the absence of cracks, pits and folds, which a roughness number alone does not capture.
Most hygienic failures are details. Product-contact joints should be continuous welds, ground and polished flush, not lap joints or spot welds that leave crevices. Metal-to-metal joints and threads in the product zone trap soil. Internal corners should be radiused; EHEDG recommends radii of at least 3 mm, with larger radii preferred. Equipment must be self-draining, with slopes and outlets placed so that no liquid pools after cleaning. Horizontal surfaces, ledges and hollow sections that are not sealed collect dirt and harbour pests.
Pipework avoids dead legs, the closed branches off a flowing line where cleaning solution barely circulates; guidelines limit their length relative to pipe diameter and favour orienting tees so that flow sweeps the branch. Valves, pumps and sensors in the product zone must be of hygienic design, cleanable in place or easily dismantled. Bearings sit outside the product zone, and shaft seals are designed to keep lubricant out of product. Equipment stands clear of floors and walls by enough distance to clean beneath and behind it, and frames use closed or round sections rather than open channels. Fasteners are minimised and never placed above open product.
Buildings follow the same logic: impervious floors laid to falls towards hygienic drains, coved joints between floor and wall, smooth washable walls, no false ceilings over open product without access for cleaning, and drains that run from high-care towards low-care areas, never the reverse.
Plant location decides costs for the life of the project. Food plants weigh factors that ordinary manufacturing does not stress as much. Perishable raw material favours a site near the growing area (tomato, sugarcane, milk collection, fish landing), because quality falls and transport cost rises with distance; products that lose weight in processing also favour raw-material sites. Products that gain weight or are perishable after processing (bread, fresh dairy) favour market sites. Other factors are a reliable supply of potable water in the volume needed, power and fuel, a way to treat and discharge effluent within legal limits, labour with the right skills, road, rail or port access, land cost and soil, climate, flood risk, zoning and local consent, and government incentives such as food parks.
Several quantitative methods support the choice. The factor-rating method gives each factor a weight summing to one, scores each site on each factor, and ranks sites by the weighted sum; it makes judgement explicit but is only as good as its weights. The centre-of-gravity method locates a facility to minimise transport by taking the load-weighted average of the coordinates of the sources or markets it serves: X = Σ(xi·Wi)/ΣWi and Y = Σ(yi·Wi)/ΣWi. Locational break-even analysis plots total cost (fixed plus variable per unit times volume) for each site against volume; the crossover volume between two sites is the difference in fixed costs divided by the difference in variable costs per unit. Transportation linear programming handles several plants and markets together.
Plant layout arranges equipment, people, materials and services to give safe, hygienic, efficient flow. Product (line) layout arranges equipment in the sequence of operations for one product or a family of similar products; it suits continuous, high-volume food lines such as milk, juice, biscuits and flour, with low handling and work-in-progress but little flexibility, and a breakdown stops the whole line. Process (functional) layout groups similar machines together; it suits small-batch, many-product plants and job shops, at the cost of more handling and longer flow paths. Fixed-position layout brings resources to a product too large to move and is rare in food. Combination or cellular layouts group the equipment for a product family into cells.
Systematic layout planning, developed by Richard Muther, is the standard procedure. It starts from product–quantity analysis and flow of materials, then rates how close each pair of activities needs to be on an activity relationship chart using the letters A (absolutely necessary), E (especially important), I (important), O (ordinary), U (unimportant) and X (undesirable). A relationship diagram is drawn, space requirements are added, and alternative block layouts are developed and evaluated. From-to charts quantify material movement between departments.
Hygiene overrides convenience. Flow should be one-directional from raw material receiving through processing to finished-goods dispatch, so that raw and finished products never cross. Receiving and dispatch are separated. Low-risk and high-care zones are physically divided, with separate staff routes, changing rooms and, where needed, separate air handling. Waste leaves by its own route. Utilities (boiler house, refrigeration plant room, compressor room, effluent treatment) sit outside production areas, with the boiler and effluent plant placed downwind of open-product areas where possible. Space is planned for cleaning around equipment, for maintenance access, for future expansion along the line, and for storage of raw materials, packaging and finished goods sized to the supply cycle. Single-storey buildings are now usual; multi-storey buildings remain in flour mills and some older plants where gravity flow saves conveying.
A layout is checked by walking it on paper: trace each material, person, piece of waste and utility from entry to exit and look for crossings, backtracking and bottlenecks.
A flow sheet is the language engineers use to describe a process. A block flow diagram shows each operation as a rectangle with arrows for the main streams; it is used in early feasibility work and for the HACCP flow diagram, which must show every step including rework, by-product and waste streams. A process flow diagram (PFD) shows the major equipment with symbols, the main process lines, and a stream table that gives, for each numbered stream, mass flow, composition, temperature and pressure; it records the material and energy balance. A piping and instrumentation diagram (P&ID) adds every line with its size and number, valves, instruments, control loops, interlocks and utility connections; it is the basis for detailed engineering, construction and hazard studies. Symbols follow standards such as ISO 10628 for diagrams and ISA-5.1 for instrumentation tags, in which the first letter gives the measured variable (T temperature, P pressure, F flow, L level) and the following letters give the function (I indicate, C control, T transmit, A alarm).
The principles of steady-state material and energy balances, and the balances for individual unit operations such as evaporators, dryers and heat exchangers, are taught in the dairy and food engineering subject. At plant-design level the same principle, input equals output at steady state, is applied to the whole line to size it. The designer starts from the required output of finished product and works backwards through each step, applying its yield (mass of output per mass of input). The overall yield of steps in series is the product of the individual yields, so a line with 85 % yield at peeling and 95 % at trimming has an overall yield of 0.85 × 0.95 = 0.8075, and the raw material needed is the finished output divided by that figure. Moisture removed, by-products and waste streams are carried in the same table, because they size the dryer, the by-product handling and the effluent plant.
Energy balances for the line add up the heating and cooling duties of each step to give the steam and refrigeration loads (chapter 11). Sensible heat is m·cp·ΔT; latent heat is m·λ. Process integration then looks for heat recovery, such as regeneration in a pasteuriser or using the vapour of an evaporator to preheat feed.
Equipment in series must be balanced. The capacity of the line is the capacity of its slowest step, the bottleneck. Other steps run below their capacity, and their utilisation is line capacity divided by their own capacity. Surge tanks or buffers between steps absorb short stoppages but do not raise long-run capacity above the bottleneck. When a step must be duplicated, the number of units is the required throughput divided by the effective capacity of one unit (rated capacity times its efficiency), rounded up.
Overall equipment effectiveness (OEE) is a common performance measure: availability (run time divided by planned time) × performance (actual rate divided by ideal rate) × quality (good output divided by total output). Food plants also lose time to cleaning, product changeover and allergen clean-downs, which must be planned in the production schedule and in sizing.
A feasibility study asks whether a proposed plant should be built; a detailed project report (DPR) is the document lenders, investors and subsidy agencies use to decide. Both normally cover market feasibility (demand, competition, prices, marketing channels), technical feasibility (product, process, capacity, raw-material availability and seasonality, location, layout, equipment, utilities, manpower, implementation schedule), financial feasibility (project cost, means of finance, cost of production, profitability, cash flow, break-even, return measures, sensitivity), and environmental, social and regulatory aspects (licences, effluent consent, food-safety approval). In agro-processing the length of the raw-material season is often the deciding factor, because capacity utilisation across the year drives cost per unit.
Project cost is fixed capital plus working capital. Fixed capital includes land and site development, buildings, plant and machinery, installation, utilities, pre-operative expenses and contingency. Early estimates scale from known costs. The six-tenths rule states that the cost of equipment or a plant varies with capacity raised to an exponent near 0.6: C2 = C1·(Q2/Q1)^0.6; the exponent differs by equipment type, so published exponents are used where available. Lang factors, given in standard process-design texts as 3.10 for solid-processing, 3.63 for solid–fluid and 4.74 for fluid-processing plants, multiply delivered equipment cost to estimate fixed capital. Working capital funds raw-material stocks, work in progress, finished goods and receivables; seasonal food plants need much more of it than year-round plants, because a season's raw material is bought in a few months.
Cost of production splits into fixed costs (depreciation, interest on term loans, insurance, salaried staff, a share of maintenance) and variable costs (raw materials, packaging, utilities, wages tied to output). The break-even quantity is fixed cost divided by contribution per unit, where contribution is selling price minus variable cost per unit; break-even capacity utilisation is that quantity divided by installed capacity. A low break-even utilisation gives a margin of safety against poor seasons.
Depreciation spreads the cost of an asset over its life. The straight-line method charges (cost − salvage value)/life each year; the declining-balance method charges a fixed percentage of the remaining book value, giving larger charges early. Tax law in each country prescribes which method and rates are allowed for tax purposes.
The payback period is the time for cumulative net cash inflows to recover the investment; it is simple and favours liquidity but ignores the time value of money and cash flows after payback. Return on investment expresses annual net profit as a percentage of investment. Discounted measures are preferred for decisions. Net present value (NPV) is the sum of each year's net cash flow divided by (1 + i)^t, minus the initial investment; a project with positive NPV at the required discount rate adds value. For a uniform annual cash flow A over n years, the present value is A × [1 − (1 + i)^−n]/i. The internal rate of return (IRR) is the discount rate at which NPV is zero; the project is acceptable if IRR exceeds the cost of capital. The benefit–cost ratio is the present value of benefits divided by the present value of costs. Lenders also examine the debt service coverage ratio (DSCR), cash available for debt service (net profit after tax plus depreciation plus interest) divided by debt service (principal repayment plus interest) for each year.
Sensitivity analysis recalculates these measures when key inputs change, such as raw-material price, selling price, capacity utilisation and project delay, and shows which assumptions the project depends on. A food project that is viable only at full utilisation in every season is fragile.
Sizing starts from the line balance. For each step the designer has a mass flow, an operating time per day and, from the yield table, the inflow and outflow. Batch equipment is sized from batch volume and cycle time (fill, process, empty, clean); continuous equipment from flow rate and residence time. Volume of a holding vessel equals flow rate times holding time; the vessel's total volume is the working volume divided by the allowed fill fraction, leaving headspace for foaming, agitation and surge. Conveyor capacity is cross-sectional load area × belt or screw speed × bulk density. A design margin (overdesign factor) is added for uncertainty, and parallel units are considered where a single failure would stop the plant or where cleaning must proceed while production continues.
Pump power is the hydraulic power ρ·g·Q·H divided by pump efficiency; the motor is then selected one standard size above. Heat exchangers, evaporators, dryers and freezers are sized by the methods in the dairy and food engineering, heat transfer and refrigeration subjects; the plant designer's task is to collect their duties into utility loads.
Steam is the main heating medium in food plants because it carries a large latent heat at a temperature fixed by its pressure, and it condenses on the heat-transfer surface. The steam demand of a step is its heat duty divided by the latent heat of steam at the supply pressure (taken from steam tables), increased for heat losses. Boilers are compared by their equivalent evaporation 'from and at 100 °C': the actual steam rate multiplied by (enthalpy of steam produced − enthalpy of feed water) and divided by the latent heat at 100 °C, 2257 kJ/kg. Boiler efficiency is the heat given to steam divided by the heat in the fuel fired (fuel rate × calorific value). Returning condensate saves fuel and treated water. Agro-processing plants often burn their own residues (husk, bagasse, shells) in suitable boilers, so fuel availability and ash handling belong in the design.
Water demand is the sum of process water (ingredient, washing, transport flumes), cleaning water, boiler feed, cooling-tower make-up and domestic use. It is estimated from specific water use per tonne of product, which varies widely with product and plant practice, and verified by metering. Storage is sized to cover interruptions of supply. Treatment depends on use: potable quality for food contact, softening and deaeration for boiler feed, and treatment against scale and biological growth for cooling water.
Refrigeration load is the sum of the product load (sensible heat above freezing, latent heat of freezing and sensible heat below freezing, plus respiration heat for live produce), transmission through walls, air infiltration through doors, and internal loads from people, lights, motors and defrost. A ton of refrigeration (TR) equals 3.517 kW, the rate of heat removal that freezes one short ton of water at 0 °C in 24 hours. Load calculation and cycle design belong to the refrigeration and cold-chain subject; plant design sizes the plant room and condensing system and provides for peak loads, such as the start of a harvest season.
Compressed air drives cylinders, valves, packaging machines and product conveying. Demand is added up as free air delivery (volume at atmospheric conditions) for each user, adjusted for how much of the time each operates, plus an allowance for leakage, and the compressor is chosen to cover it with a receiver to smooth peaks. Air that contacts food or food-contact surfaces must be filtered and free of oil and water to a stated quality class; ISO 8573-1 defines such classes. Electrical design adds the connected load of all motors and heaters and applies a demand (diversity) factor, because not every load runs at full power at once, to set the transformer and standby generator size.
Food-processing wastewater is rich in dissolved and suspended organic matter: sugars, starch, protein, fat, oil and grease (FOG), with fluctuating pH from cleaning chemicals and high temperatures. Biochemical oxygen demand (BOD) is the oxygen consumed by microorganisms decomposing the organic matter, measured over five days at 20 °C (BOD5). Chemical oxygen demand (COD) is the oxygen equivalent of organic matter oxidised by a strong chemical oxidant; it is measured in hours and is always at least as large as BOD. A high BOD/COD ratio indicates waste that is readily biodegradable, which most food waste is. Other parameters are total suspended solids, FOG, total nitrogen and phosphorus, pH and temperature. Pollution load in kg per day is concentration (mg/L, which equals g/m³) times flow (m³/day) divided by 1000.
Discharge limits are set by each country, and often each state or municipality, and differ for discharge to a sewer, to surface water or to land. The design starts by measuring flow and load through the production season, because food effluent varies hour by hour and season by season.
The first step is always to reduce waste at source: dry clean-up before washing, recovery of product from pipes and vessels before rinsing, segregation of strong streams, and water reuse where hygiene allows. Treatment then proceeds in stages. Preliminary and primary treatment remove coarse solids by screens, grit, fat and oil by gravity traps or dissolved-air flotation, and settleable solids in clarifiers; an equalisation tank evens out flow and pH. Secondary treatment removes dissolved organic matter biologically. Aerobic systems include activated sludge, sequencing batch reactors and trickling filters. Anaerobic systems, such as the upflow anaerobic sludge blanket (UASB) reactor and covered lagoons, suit strong food effluents: they produce less sludge, need no aeration energy and produce biogas, but usually need aerobic polishing to meet discharge limits. Tertiary treatment (filtration, disinfection, nutrient removal) is added where reuse or strict limits demand it. Sludge is thickened, dewatered and composted, digested or disposed of under local rules. Solid wastes follow the waste hierarchy: prevention, reuse as food or feed, recycling, recovery of materials or energy, and disposal last.
A by-product is a secondary product with a market; turning waste into by-products improves both the environment and the plant's economics. Classic examples: rice bran yields edible oil and de-oiled bran for feed, and rice husk is a boiler fuel whose ash is rich in silica; wheat bran and germ go to feed and food; sugarcane bagasse fuels cogeneration in sugar mills and molasses is fermented to ethanol; whey from cheese and paneer is concentrated or dried, or fractionated by membranes into whey protein concentrate and lactose; citrus peel and apple pomace are sources of pectin, essential oils and dietary fibre; fruit and vegetable trimmings become feed, compost or biogas; fish and meat by-products are rendered into meal, oil and gelatin. Mango and other seed kernels, coconut shell (activated carbon), groundnut shells and coffee pulp have their own uses in producing regions.
The engineer judges a by-product line like any other investment: is the quantity steady enough, can the material be stabilised before it spoils (many wet by-products ferment within hours), is there a market within reach, and does the value cover collection, processing and quality control? The by-product must also be handled hygienically if it enters the food or feed chain, which brings it inside the plant's food-safety system.
A B.Tech student uses this subject to read and draw flow sheets, compute control-chart limits and capability indices from plant data, write a HACCP plan for a simple product as a class project, lay out a small processing unit with one-way flow and correct hygiene zoning, and prepare the financial chapter of a project report with break-even, payback and NPV. Internships in processing plants test exactly these skills: students are often handed a filler's weight records or a line's downtime log and asked what is wrong.
An engineer five years into a processing plant runs utilities and projects. Typical tasks are sizing a new boiler or chiller after a capacity expansion, writing the engineering parts of the food safety plan, choosing hygienic valves and pumps for a new line, leading root-cause analysis when an X̄ chart drifts or a metal detector fails a test, preparing the DPR for a subsidy or bank loan, and making the effluent plant meet consent limits during the peak season. Decisions involve trade-offs: a cheaper machine that is hard to clean, or a layout that saves floor area but crosses raw and cooked flows.
A plant or projects head with twenty-five years' experience decides where and what to build. They compare sites with factor rating and cost models, set the capacity against the raw-material season, approve layouts for food-safety audits and export certification, and judge whether a by-product line (bran oil, whey protein, pectin, biogas) earns its capital. They represent the plant before regulators and buyers' auditors, decide what to do with product after a deviation, and design recall plans. Their work is to balance capital cost, operating cost, safety and compliance across several countries' rules.
A researcher with a doctorate and long industry experience works on the questions the standards leave open: validating process controls for new technologies, building risk models that link sampling plans to public-health outcomes, designing equipment surfaces and geometries that resist biofilm, developing rapid and non-destructive quality measurements, and life-cycle and techno-economic assessment of by-product valorisation. They also serve on standards and Codex expert committees, write guidance for regulators, and train the next generation of plant engineers and food-safety auditors.
This manual does not give national discharge limits, FSSAI licensing turnover thresholds, specific microbiological criteria (n, c, m, M values) or detergent and sanitiser concentrations, because these differ by country and product and change often; readers must take them from the current national text. Dead-leg length limits and minimum radii are described in principle, not quoted from every guideline. Numerical examples of steam, refrigeration and water use per tonne of product are not given because published figures vary widely. Detailed refrigeration load calculation, unit-operation design and clean-in-place are covered in other subjects. Food defence, allergen labelling law and traceability rules are introduced only briefly.
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.
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.