🍞 Production Planning, Equipment and Premises

A bakery day is not a recipe, it is a timetable. The recipe tells you what goes in the bowl. The timetable tells you whether the bread reaches the customer while it is still worth buying, and it is built backwards from that hour, never forwards from when you feel like starting.

Start at the end. Write down the hour your first customer must have product in hand. Everything else is subtraction.

Work backwards through the fixed stages. Some of these have sourced times and some you must measure yourself.

1. Delivery and display. Your own travel time.
2. Wrapping. Bread must not be wrapped above 35 degrees Celsius, because wrapping warm bread traps moisture in the bag and makes a mould incubator. It also must not be sliced above that temperature or it caves in.
3. Cooling. Typical cooling to that wrapping temperature takes two to three hours. This is dead time you cannot compress by wishing, and it is where most beginners lose their morning.
4. Baking. Your own oven, your own times, measured with a probe thermometer rather than a clock. Lean-dough bread finishes at an internal temperature of 88 to 99 degrees Celsius; rich doughs such as sandwich bread and rolls at 82 to 88; scones and quick breads at 93 to 96.
5. Final proof. Published figures give a recommended proofing temperature of 35 to 37 degrees Celsius at 85 to 95 percent relative humidity, with a standard duration of 60 to 65 minutes. Short-mix doughs take up to an hour; improved and intensive-mix doughs one to two hours; sourdough-leavened doughs longer than two hours.
6. Moulding and intermediate rest. Measure your own.
7. Bulk fermentation. No technical source gives reliable time bands for this, and this course will not invent one. Ferment to condition, not to the clock, and use dough temperature control to make your own timings repeatable from day to day.
8. Mixing. One published description of a straight-dough process gives 12 to 18 minutes of high-speed mixing in a horizontal mixer. Your mixer is not that mixer, so time your own.
9. Weighing and preparation.

Two published overall timelines anchor the whole plan. A straight-dough process runs about 3 to 4 hours from scaling through to packaging. A sponge-and-dough process runs 6 to 8 hours. That is the single most important planning fact in this lesson: choosing a process is choosing what time you get out of bed.

Work a real timetable. Customer needs bread at 07:00. Travel is 30 minutes. Using the straight-dough figures:

• 06:30 leave the bakery.
• 06:15 wrapping finished; loaf must be at or below 35 degrees Celsius.
• 03:45 out of the oven, allowing two and a half hours of cooling.
• 03:05 into the oven, allowing 40 minutes bake, measured by probe.
• 02:00 final proof begins, allowing 65 minutes.
• 01:30 moulding and panning.
• 01:00 divide and rest.
• 23:45 mix, 15 minutes, then bulk ferment to condition.
• 23:15 weigh out.

That is a start before midnight for bread at seven, and it is why bakers keep the hours they keep. Look at it and you can see your options: a sponge process would push the start three or four hours earlier still, while cutting the cooling short is not an option because it breaks the wrapping rule.

Now sequence more than one product. One oven can only do one thing at a time, so the rule is to load it continuously and never let it stand hot and empty, because an empty oven still burns fuel and still carries overhead. Bake in order of temperature, hottest first, working down as the oven falls, and put the products that tolerate a falling oven at the end. Biscuits are useful last because they absorb residual heat and become stock that keeps.

There is a second sequencing rule that is a safety rule, not an efficiency one. Where allergens are involved, process the foods without an allergen before the foods with it, add allergenic ingredients as late in the process as possible, and run products with the same allergen profile one after another. In a one-room bakery you cannot separate by walls, so you separate by time and by cleaning between runs.

Finally, write the timetable on the wall with clock times, not durations. A plan in durations is a plan you have to recalculate at three in the morning, which means it is a plan you will abandon.

The order in which you buy equipment matters more than the total you spend. Most failed bakeries did not buy too little; they bought in the wrong order, spending the working capital on the impressive machine and leaving nothing for the flour.

Buy in this order.

First, an accurate scale. Baker's percentage is unusable without one, and so is every costing in this course. If you cannot weigh, you cannot repeat, and if you cannot repeat you have no product, only a series of accidents.

Second, a probe thermometer. It is the cheapest quality-control instrument in the bakery and three separate controls depend on it: dough temperature at the end of mixing, internal doneness at the end of the bake, and the 35 degrees Celsius ceiling for slicing and wrapping. A baker with a thermometer can diagnose problems. A baker without one is guessing.

Third, the oven, because it sets your capacity, your fuel bill and your working hours. The tiers, with indicative capital figures that must be verified locally, run like this:

1. Drum or improvised wood oven. Lowest capital, highest fuel use, poorest temperature control. No capital figure was available to this course, so price one from a local fabricator.
2. Improved wood-fired stove or oven. The Mirt stove in Ethiopia is quoted at about USD 11 for a single unit and about USD 126 for a clustered four-unit installation, with about 50 percent fuel reduction against an open fire.
3. Improved or rocket baking oven. Ugandan rocket ovens are quoted at about USD 1,616 to USD 7,474, with capacity from 12 to 24 loaves at the small end up to 760 at the large end, an efficiency increase of about 70 percent, preheat time cut by at least two thirds, and firewood falling to about one tenth of previous use. These ovens bake at about 250 degrees Celsius.
4. Electric deck oven, two to three decks, indicatively R9,000 to R30,000; gas deck oven of the same size, indicatively R25,000 to R55,000.
5. Commercial rotary or reel oven, indicatively R200,000 to R380,000.

Every currency figure there comes from a single commercial guide or a development wiki, for one country at one date. Duty, freight and tax dominate the delivered price in most African markets, and a locally fabricated or second-hand unit can cost an order of magnitude less than a new import. Treat them as the shape of the ladder, not as prices, and get three local quotations before you decide anything.

Fourth, a mixer, once your arms are the bottleneck rather than your oven. A spiral dough mixer of 20 to 40 litres is quoted indicatively at R12,000 to R28,000. Before you buy, do the arithmetic in the other direction: how many extra saleable units per day will it produce, times your contribution per unit, times trading days, against the monthly repayment plus the added depreciation. If that sum does not clear comfortably, the mixer is a want and not yet a need.

Fifth, and only fifth, everything else. A proofer or retarder is quoted indicatively at R8,000 to R18,000. Costs for scales, dough dividers, moulders, slicers, prep tables, racks and packaging equipment were not available to this course at all, so you must price them locally. That gap is a reminder rather than a problem: local prices are the only prices that matter, and every one of these items can be quoted by a supplier in your own town in an afternoon.

Three principles cut across the whole list.

Buy capacity you can fill. An oven running half empty spreads the same overhead over fewer loaves, and that cost never appears on an invoice. Match the oven to loaves you can genuinely sell, not to the bakery you imagine.

Buy what can be repaired where you live. A machine with no local spare parts and no local technician is a machine that will one day stop your bakery for a month. Ask, before buying, who repairs this and where the parts come from.

Buy fuel efficiency deliberately, with the payback calculated. One documented Ethiopian case recorded annual fuel cost falling from USD 1,900 to USD 970, a saving of USD 930 a year, with income up USD 730, and payback of about 15 months. Do the same sum on your own quotation: capital divided by annual saving plus annual extra income, times twelve, gives payback in months.

Last, remember that food-contact surfaces must be smooth, non-absorbent and inert to food, detergents and disinfectants. A cheap bench that soaks up water is not a saving.

A small bakery is usually one room, and one room is not an excuse for chaos. Layout is the cheapest productivity tool you will ever buy, because it costs nothing but thought, and a bad layout charges you a few wasted minutes on every single batch for years.

The governing idea is one-direction flow. Raw materials enter at one point, move through the process without ever going backwards, and leave as wrapped product at another point. Draw your room on paper and mark, in order: goods in, dry store, weighing, mixing, bench and dividing, proofing, oven, cooling, wrapping, finished store, goods out. Then draw the path a bag of flour takes. If that path crosses itself, or if wrapped bread has to pass back through the flour area, you have found the problem before it cost you anything.

Why does crossing matter? Because everything that spoils bread arrives after the oven. Baking destroys vegetative mould, so every mould spore on a loaf landed there afterwards, from unclean equipment, contaminated wrappers, mouldy racks and tools, or dust. Flour dust drifting over cooling loaves is not a small matter; it is the transport system for the next month of complaints. Keep the cooling and wrapping end clean, still and away from the dusty end.

Now the fixed requirements, drawn from recognised food-hygiene principles. Treat these as the floor, and confirm the detail with your own local environmental health office, because premises requirements are national and often municipal.

• Food-contact surfaces must be smooth, non-absorbent and inert to the food, to detergents and to disinfectants. Bare timber benches fail this on the first count.
• There must be adequate means of washing and drying hands, including soap, preferably liquid, and wash basins supplied with hot and cold water.
• The hand-wash basin cannot be used for any other purpose. Not for washing food, not for washing utensils. That means a small bakery needs at least two water points, not one.
• Cleaning facilities must have an adequate supply of hot and cold water.
• Waste containers must be specifically identifiable, suitably constructed, and made of impervious material.
• Premises must be designed to give effective protection against pest access and harbourage, and the site itself should not be one prone to infestation.
• Storage must protect food from contamination, including allergen cross-contact.

For storage, the practical rules follow from pest control: keep raw materials off the floor and away from walls, rotate stock, remove waste and spillage every day, keep a monitoring record, and never use agricultural pesticides in a food area. Which rodenticides are permitted, and what a pest-control programme must contain, is set nationally, so ask.

Now the hardest part of small-bakery layout, which is allergens. A bakery is an allergen-dense place by nature: wheat, egg, milk, soy in shortening and lecithin, nuts and sesame are all routine. The recognised controls are physical separation by walls and partitions, or separation in time with suitable cleaning and disinfection between uses. In one room, physical separation is usually impossible, so separation in time is the only workable route, and it must be done properly:

1. Schedule products without an allergen before products with it.
2. Add allergenic ingredients as late in the process as possible.
3. Run products with identical allergen profiles consecutively.
4. Clean between runs and confirm the equipment is visibly clean.
5. Store rework in sturdy, covered, clearly marked containers in a designated area, and return rework to the same product wherever you can.
6. Check the right label goes on the right product, and have a way to segregate and re-label or destroy anything mislabelled.

Note what is not on that list: a "may contain" statement. Precautionary labelling is not a substitute for control, and whether it is permitted at all, and in what words, is a national matter. A bakery that can eliminate cross-contact should do that rather than label around it.

One workflow habit is worth as much as the layout. Set up the next stage before you finish the current one: tins greased before dividing, racks placed before the oven opens, wrappers counted before cooling ends. In a small bakery the oven waits for nobody, and the cost of a five-minute delay at the oven door is paid in fuel, in overhead and in a queue that builds all morning.

Flour is usually the largest single line in a bakery's cost, and it is the ingredient most bakers buy blind: whatever the shop has, at whatever price, in whatever condition. A written purchase specification changes the relationship. It tells the miller you know what you are buying, and it gives you something to point at when a delivery bakes badly.

Ask for a certificate of analysis, usually called a CoA, and specify these things.

Protein. Bread flour runs at 12 to 14 percent protein, all-purpose at 8 to 11, pastry flour 8 to 9, cake flour 7 to 9, and biscuit flour 8 to 10. The Codex minimum for wheat flour is 7.0 percent on a dry-weight basis, but that is a floor for the commodity, not a bread specification, and a 7 percent flour will not make good bread. Note also that quantity is not quality: a high-protein flour with weak gluten still bakes badly, and gluten quality is measured on an extensograph or alveograph, not by the protein number.

Falling Number. This is the single most useful specification a small baker can learn to ask for. It measures enzyme activity, in seconds, and low seconds means high activity. Below 220 seconds usually means sprouted wheat: sticky dough that is hard to handle, large open holes, sticky crumb that fouls the slicer, and an over-dark crust. Over 250 seconds is described as suitable for most breadmaking, with 250 to 280 stated as the optimal band for large-scale bread baking. Above 300 seconds means minimal enzyme activity: long proofs, pale crust, low volume and dense crumb. Above 350 seconds usually needs enzyme supplementation. Real flours run from about 55 seconds up into the high 400s.

This matters enormously where rain falls at harvest, because rain causes sprouting in the field and sprouting crashes the Falling Number. A baker whose dough suddenly turns sticky and whose loaves come out dark has very probably bought a low Falling Number flour rather than made a mistake.

Ash. Typical commercial flours sit below 0.6 percent, and patent flour can be as low as 0.35 percent. Codex sets no maximum for ash at all, calling it buyer preference. That is the lesson: ash is a commercial specification you negotiate with your miller, not a legal one.

Moisture and safety. Codex sets a maximum of 15.5 percent moisture for wheat flour; typical flour as milled runs 14 to 15 percent, and storage problems begin above 16 percent. Codex sets a maximum of 5 micrograms per kilogram of ochratoxin A in raw wheat, barley and rye, and requires flour to be free from heavy metals, parasites and micro-organisms at hazardous amounts. Particle size is specified as at least 98 percent passing a 212 micrometre sieve.

One serious gap you must close yourself. Maximum levels for deoxynivalenol, fumonisin and aflatoxin in cereals and flour were not available to this course. For an African bakery using maize or groundnut ingredients this is a real risk, not a paperwork matter. Ask your national food safety authority for the current national limits, and ask your miller what they test for.

Now storage. Buy the best flour available and store it badly and you have wasted the money. Keep it dry, because problems begin above 16 percent moisture. Keep it off the floor and away from the walls, which is a pest-control requirement as much as a damp one. Rotate stock first in, first out, and write the delivery date on every bag as it arrives. Record the miller, the lot number printed on the bag and the delivery date, because without that link to your production dates you cannot recall a batch.

Finally, the arithmetic of stock. Two figures run your store: usage per day and cover in days.

A bakery uses 10.23 kg of flour per bake and does three bakes a day, six days a week. Daily usage is 10.23 times 3 = 30.7 kg. Weekly usage is 30.7 times 6 = 184 kg.

If 250 kg is on hand, cover is 250 divided by 30.7 = 8.1 trading days.

Set the reorder point from lead time plus a safety buffer. If your miller takes four days to deliver and you want three days of buffer, reorder at seven days of cover, which is 7 times 30.7 = 215 kg. When the store falls to 215 kg, you order, and you do not wait for a better price.

Hold too little and you stop baking. Hold too much and your cash is sitting in sacks, ageing, attracting weevils and gaining moisture. Cover in days is the number that keeps both errors visible.

In a small bakery, staffing is not a human resources exercise. It is the question of who does what, at what hour, and who checks that it was done. Get that on paper and a two-person bakery can outperform a sloppy five-person one.

Start from the timetable, not from job titles. Take your clock-time plan from Lesson 1 and write a name beside every line. You will immediately see two things: the moments where one person must be in two places, and the long gaps where somebody is being paid to wait. Both are fixable, usually by moving a preparation task into a waiting period.

A typical small-bakery split looks like this. One person owns the dough: weighing, mixing, dough temperature, dividing and moulding. One person owns the heat and the finish: oven loading, doneness testing, cooling, wrapping and counting. Selling and delivery are a third role that often falls to whoever is free, which is exactly why it gets done badly. Give it a name too.

Every person handling food carries obligations. Personnel must be competent, and management must ensure appropriate training and supervision. An appropriate degree of personal hygiene must be maintained so that people are not themselves a source of contamination. Those are principles, not detail, and the detail is national.

So here is what you must ask rather than assume. Whether food handlers in your country need a medical certificate or a food handler's permit, which symptoms or diagnoses require someone to be kept away from food, for how long, and whether a doctor's note is compulsory, is set by your national authority. In several African jurisdictions a food handler's medical certificate is a licensing condition. Ask your environmental health office directly, and write down the answer with the officer's name.

What you can teach safely, because it follows from the hygiene principles, is a hand-washing rule everybody in the bakery learns on day one. Wash on entering the production area. Wash after every toilet visit. Wash after handling raw egg or waste. Wash after touching your face, your hair or your phone. Wash between tasks with different allergen profiles. And dry your hands, because a wet hand recontaminates whatever it touches next.

Training a new person is a short list, and it is not about recipes. It is about the four measurements that make the bakery repeatable: dough temperature at the end of mixing, internal temperature at the end of the bake, temperature before slicing and wrapping, which must not exceed 35 degrees Celsius, and the weight of the scaled dough piece. Teach those four with the probe and the scale in the person's own hand, and make them sign the check sheet themselves. A check signed by the person who did the work is worth ten checks signed by the owner afterwards.

Now the money. Labour is costed the same way as everything else in this course: per unit.

Two people work a six-hour shift and produce 180 saleable loaves. If the total wage cost of that shift, including your own time at the rate you would have to pay a replacement, is 60 CU as an illustrative figure, then labour per loaf is 60 divided by 180 = 0.33 CU. Substitute your own wage rates; these are illustrative figures shown so the arithmetic is visible.

Now test hiring. Suppose an extra person costs 30 CU a shift and lets you produce 90 more loaves, which you can genuinely sell. If contribution per loaf before labour is 0.90 CU, the extra loaves bring 90 times 0.90 = 81 CU against a cost of 30 CU. That hire pays. But if you can only sell 25 more loaves, the extra brings 25 times 0.90 = 22.50 CU against 30 CU, and the hire loses money every day it exists.

That arithmetic answers the question small bakers agonise over most. Hiring is never justified by how tired you are. It is justified by units you can sell, times contribution, against the wage. Do the sum before the conversation, not after.

One last point about the people. The four record sheets, the check signatures and the timetable are not bureaucracy; they are how a bakery keeps running on the day the owner is ill. A business that only works when one particular person is present is not a business yet, it is a job with a mixer.

Growth breaks more small bakeries than shortage does. The pattern is always the same: sales rise, the baker buys capacity, quality slips, the new customers leave, and the bakery is left with the repayments. Growing well is a discipline, and it has three parts: find the real constraint, rebuild control after every change, and grow in reversible steps.

First, find the constraint. Most bakers assume it is the oven. Often it is not. Walk your own timetable and ask where the queue forms. If dough waits for oven space, the oven is the constraint. If the oven waits for dough, the mixer or the bench is. If wrapped bread waits for a vehicle, distribution is the constraint and a bigger oven will only produce more bread to sit around. If your cooling racks are full at the moment you need to wrap, then racks, not capacity, are what is costing you the morning. Buying capacity anywhere except the constraint adds cost and adds nothing else.

Second, and this is the part that damages quality, understand that every change of scale destroys your process settings and you must rebuild them.

Mixing puts energy into dough as heat. The size of that heating effect is called the friction factor, and it is specific to your mixer, your batch size and your mixing time. Published values illustrate how wide the spread is: hand kneading around 1 to 3 degrees Celsius, and stand mixers around 8 to 16 degrees Celsius. Friction factors for commercial spiral and planetary mixers were not available to this course, and that is precisely why you must measure your own rather than borrowing anybody's.

Measure it with one batch and a thermometer. Friction factor equals final dough temperature times three, minus the sum of water temperature, room temperature and flour temperature. Then use it in reverse to set your water:

Water temperature = (desired dough temperature times 3) minus (flour temperature plus room temperature plus friction factor).

Work the example that matters in a hot climate. Target dough temperature 25 degrees Celsius, flour at 28, room at 30, measured friction factor 8. Water temperature = 75 minus 66 = 9 degrees Celsius. That is chilled water, most of the year. Scale up to a bigger mixer with a higher friction factor and the required water temperature drops further, which is the hidden cost of the new machine nobody quotes you.

Why does this matter for growth? Because dough temperature is the master variable that controls fermentation. Wheat yeast breads are given a target of 24 to 26 degrees Celsius for best flavour and rise, while a fast commercial straight-dough process runs 28 to 30 at the end of mixing, and 24 to 28 is given as optimum for keeping quality. Higher dough temperature means a faster process, less flavour development and poorer keeping quality. When a bakery grows and the bread suddenly does not keep as well, the cause is very often a dough temperature that crept up with the new mixer and the busier room.

Heat brings a second growth risk that is not about quality but about safety. Rope spoilage comes from Bacillus spores that arrive in the flour and survive baking, since they withstand up to 130 degrees Celsius while your loaf core reaches only about 90 to 97. It shows first as a sweetish, fruity or rotten smell in the crumb at 12 to 24 hours, then sticky crumb drawing thread-like strands. It is favoured by warm storage, crumb moisture of 35 to 42 percent and water activity of 0.90 to 0.95, and it is markedly more frequent in hot seasons. Every one of the mistakes growth encourages makes rope more likely: stacking loaves to cool because there is no rack space, wrapping before things are cool enough, and taking back returned bread into the production area. Cool rapidly, do not stack during cooling, disinfect the slicer, and eliminate contaminated leftover bread from the cycle entirely.

Third, grow in reversible steps. Before committing capital, run the extra volume for a month using overtime, an extra bake in the existing oven, or a hired hand, and see whether the market really absorbs it. If it does, buy. If it does not, you have lost a month of overtime rather than three years of repayments. Remember that fixed costs move in jumps: a second employee, a bigger room, another delivery run, and each jump raises your break-even before it raises your income.

One final rule. Do not grow your product range and your volume at the same time. Each new line brings a new ingredient, new scaling weights, new oven settings and a new wastage pattern. Add volume on the lines you already make well, get control back, and only then add a line.