foratvise
Get Started
Forming, baking and cooking

Forming, moulding and depositing

Volumetric dividing, rounding and moulding of bread dough

Validated modelOEE bound 0.958 [0.938–0.978]20,000 Monte Carlo samples · seed 20261115 · computed November 16, 2026

In short

Forming, moulding and depositing can use at most 96 of every 100 planned hours. The rest goes to limits no plant can avoid.

  1. 96usable at most
  2. 4are a physical limit

The three limits that set it

Running time: up to 96 %

The rest are stops imposed by physics or by regulation: cleaning, changeovers, defrosting.

Speed: up to 100 %

How close to its maximum speed it can run without damaging the product or the equipment.

Good product: up to 100 %

Some product falls outside specification even with the process well under control.

See the technical detail

Technical detail

Uncertainty intervals, physical models per station and sources, for anyone who wants to check each figure.

1Scenario

Mixed wheat dough is tipped into the hopper of a volumetric divider whose pockets cut pieces of fixed volume; the pieces are rounded, rested and moulded into pans and then proofed and baked into prepackaged loaves of 800 g nominal weight (the loaf of the baking model). The divider meters volume, not mass: the piece mass scatters with the divider and drifts as the dough keeps fermenting in the hopper while a batch is divided. The legal rules for prepackages fix the smallest mean dough mass that keeps the loaves legal; pieces that would make a loaf short by more than twice the tolerable error are removed at the check-weigher and returned to the hopper. The same ceilings are applied as an approximation to the depositing and moulding of other doughs and batters sold by weight (derivation, section 7).

Provenance of inputsComputed1Declared assumption11Verified quote5

2Results

Point value from the nominal parameters and 95% interval from 20,000 Monte Carlo draws of the uncertain parameters.
  • TheoreticalThermodynamic minimum energy per kg of dough (no chemical change, deformation work recoverable in the reversible limit)0 kJ/kg
  • Time to divide one hopper batch at the installed rate3.37 [1.52–6.82] min
  • Standard deviation of the piece mass from the density drift during a batch0.866 [0.326–2.715] g
  • Standard deviation of the dough piece mass (divider scatter and density drift)6.06 [3.35–9.91] g
  • Standard deviation of the loaf weight carried over from forming5.48 [3.02–9.00] g
  • Smallest legal mean loaf weight for the forming scatter804.7 [802.5–806.9] g
  • Target mean mass of the dough piece891 [855–928] g
  • Overweight given away above the nominal quantity0.00582 [0.00307–0.00862]
  • Share of loaves with a deficiency beyond the tolerable negative error (allowed up to 2.5 %)1.66 × 10⁻⁴ [2.29 × 10⁻¹¹–1.86 × 10⁻²]
  • Share of dough pieces rejected at the check-weigher (loaf would have a T2 error)1.24 × 10⁻¹⁰ [9.27 × 10⁻³¹–7.67 × 10⁻⁵]
  • TheoreticalMaximum OEE quality factor1.000000 [0.999923–1.000000]
  • TheoreticalMaximum yield: nominal quantity sold per mean quantity made (give-away; rejected dough is reworked)0.9942 [0.9915–0.9969]
  • Dough pieces per hour of the installed divider (declared, no ceiling)6,000 [3,713–8,283] 1/h
  • Dough throughput of the installed divider (declared, no ceiling)5,345 [3,312–7,415] kg/h
  • TheoreticalMaximum availability (daily cleaning of divider, rounder and moulder)0.958 [0.938–0.978]
  • TheoreticalMaximum performance factor (no physical rate limit binding)1.00
  • TheoreticalPhysics-bounded OEE0.958 [0.938–0.978]

3Ceilings per dimension

Each dimension shows the ceilings the model computes and, where units are comparable, documented benchmarks. Colour and marker shape identify the ceiling type throughout the Oracle.
  • Theoretical
  • Best available technique
  • Best documented
  • Typical
Energy per unit of productkJ/kg · lower is better
  • TheoreticalThermodynamic minimum energy per kg of dough (no chemical change, deformation work recoverable in the reversible limit)0 kJ/kg
  • TheoreticalDOE bandwidth study: products without significant chemical reactions have a zero thermodynamic minimum (dough forming is not covered by the study; the principle applies) · Benchmark0 kJ/kg
Qualityhigher is better
  • TheoreticalMaximum OEE quality factor1.000000 [0.999923–1.000000]
Yieldhigher is better
  • TheoreticalMaximum yield: nominal quantity sold per mean quantity made (give-away; rejected dough is reworked)0.9942 [0.9915–0.9969]
Availabilityhigher is better
  • TheoreticalMaximum availability (daily cleaning of divider, rounder and moulder)0.958 [0.938–0.978]
Performancehigher is better
  • TheoreticalMaximum performance factor (no physical rate limit binding)1.00
OEEhigher is better
  • TheoreticalPhysics-bounded OEE0.958 [0.938–0.978]

4Loss cascade

What separates each ceiling from the next one, from the physical limit to typical operation.
  1. TheoreticalTypicalQuality

    With the declared divider scatter the legal mean is set by the average requirement, and only the far tail of the dough-piece distribution would make a loaf with a T2 error, so almost no piece is rejected. A worse divider, a slow divider that lets the dough ferment longer in the hopper, or a large batch raise the scatter: once the scatter exceeds the tolerable error divided by 1.96 the 2.5 % rule for T1 errors sets the mean, the give-away grows linearly with the scatter and rejects start. Crippled and double pieces, and pieces torn by sticking dough, are further technology losses.

  2. TheoreticalTypicalYield

    The give-away is set by the lot-mean uncertainty and, for large scatter, by the T1 rule; it is product sold above the nominal quantity, a yield loss and not an OEE loss. Rejected pieces are reworked in the next batch and do not reduce the yield, although rework changes the dough and is limited in practice.

5Physics-bounded OEE

The highest OEE the operation can reach: availability, performance and quality limited only by unavoidable stops and physical capability.
Fraction of loading time that remains after each unavoidable loss. Axis starts at 0.90.
  • Loading time (ideal)
    1.000
  • Availability
    factor 0.958 [0.938–0.978]
    0.958 (−0.042)
  • Performance
    factor 1.00
    0.958 (−0.000)
  • Quality
    factor 1.000000 [0.999923–1.000000]
    0.958 (−0.000)
  • Physics-bounded OEE
    0.958 [0.938–0.978]

6Full derivation

Mass and energy balances, rate-limiting step, process capability and unavoidable stops, with every term explained.

Scope. Mixed wheat dough is tipped into the hopper of a volumetric divider whose pockets cut pieces of fixed volume; the pieces are rounded, rested and moulded into pans, then proofed and baked into prepackaged loaves of 800 g nominal weight, the loaf of the baking model. This is the operation forming of the bread process. Symbols, values, units, uncertainty distributions and sources are listed in the parameter table generated from config/physics/forming.yaml.

1. Mass balance — dough-piece mass, legal target and maximum yield

What the divider meters. A volumetric divider fills a pocket of volume VV and cuts it off, so the piece mass is m=ρVm = \rho V: it scatters with the fill of the pocket and with the density of the dough. Two contributions are separated.

  • The divider scatter at constant density, σdiv\sigma_{div} (declared 6 g, the same value and range as baking.sigma_divider, so that both models describe the same machine).
  • The density drift. The yeast keeps producing gas while a mixer batch of mass MM waits in the hopper, so the density falls during the dividing time of the batch at a relative rate rρr_\rho (declared 0.1 % per minute). The divider is set for the batch, so the piece mass drifts uniformly from the start to the end of the batch. The dividing time is tb=M/(Nmˉ)t_b = M/(N\bar m), where NN is the number of pieces per minute, and the drift over the batch is mˉ rρ tb=rρM/N\bar m\,r_\rho\,t_b = r_\rho M/N: the target mass cancels, and the drift depends only on the batch size and the divider rate. A uniform drift of that range has the variance (rρM/N)2/12(r_\rho M/N)^2/12.

The scatter of the dough piece and of the loaf it becomes are

σm2=σdiv2+(rρM/N)212,σw=Y σm,\sigma_m^2 = \sigma_{div}^2 + \frac{(r_\rho M/N)^2}{12}, \qquad \sigma_w = Y\,\sigma_m,

where YY is the bake yield, loaf mass per dough mass, linked from the output yield_max of the baking model so that the two models use the same dough and bread moisture. The scatter of the bake loss itself belongs to the baking model and is not counted here.

Legal target. The loaf is a prepackage. Directive 76/211/EEC and OIML R 87 state that the average content must not be less than the nominal quantity QnQ_n, that a lot may contain no more than 2.5 % of packages short by more than the tolerable negative error TT (T1 errors), and that no package may be short by more than 2T2T (T2 errors); for 500 g to 1 000 g, TT = 15 g. Following the filling model, the average requirement is met with a safety factor zμz_\mu = 2.576 on the uncertainty of the lot mean σμ\sigma_\mu (declared 2 g of dough: setting resolution of the pocket volume and density differences between batches), so the smallest legal mean loaf weight is

wˉ=max⁡ ⁣(Qn+zμ Yσμ,  Qn−T+z0.975 σw),z0.975=Φ−1(1−0.025)=1.96,\bar w = \max\!\Big(Q_n + z_\mu\,Y\sigma_\mu,\; Q_n - T + z_{0.975}\,\sigma_w\Big), \qquad z_{0.975} = \Phi^{-1}(1 - 0.025) = 1.96,

and the target dough mass is mˉ=wˉ/Y\bar m = \bar w / Y. With the point values the drift adds 0.87 g to the divider scatter, σm\sigma_m = 6.06 g, σw\sigma_w = 5.48 g, the average requirement binds and wˉ\bar w = 804.65 g, i.e. a dough piece of 890.9 g. The T1 rule takes over when σw\sigma_w exceeds (T+zμYσμ)/1.96(T + z_\mu Y \sigma_\mu)/1.96, about 10 g on the loaf.

Maximum yield. The give-away (wˉ−Qn)/Qn(\bar w - Q_n)/Q_n, 0.58 % with the point values, is product sold above the nominal quantity. As in the filling model it is a yield loss, never an OEE loss:

Ymax=Qnwˉ=0.9942.Y_{max} = \frac{Q_n}{\bar w} = 0.9942 .

Rejected pieces (section 4) are returned to the hopper and reworked into the next batch, so they do not reduce the yield.

2. Energy balance — minimum energy per unit of product

Dividing, rounding and moulding change no chemical bond: the work done on the dough deforms a viscoelastic material and is dissipated as heat or stored for a short time in the stretched gluten network, which relaxes during the rest. In the reversible limit nothing has to be spent, so the thermodynamic minimum is zero, in line with the principle of the US DOE bandwidth study that products without significant chemical reactions have a zero thermodynamic minimum (dough forming is not covered by that study). No open source documents the energy of dividers or moulders, so no best-available or typical energy level is given; the machines' electricity is small compared with mixing and baking.

3. Rate-limiting step

The divider cuts N=npfsN = n_p f_s pieces per minute (npn_p pockets, fsf_s strokes per minute; declared 4 pockets and 25 strokes per minute, 6 000 pieces per hour, about 5.3 t of dough per hour). No physical property of the dough sets a binding limit on the stroke rate in the open sources, so the installed rate is the reference of the performance factor, Pmax=1P_{max} = 1, and the rate outputs carry no ceiling. The rate does enter the physics through the density drift: a faster divider empties the hopper sooner and lowers the scatter (section 1); with the point values a batch of 300 kg is divided in 3.37 min.

4. Process capability — quality ceiling

A check-weigher after the divider removes the pieces that would become a loaf with a T2 error and returns them to the hopper. Their share is the quality loss counted here:

Qmax=1−Φ ⁣(Qn−2T−wˉσw).Q_{max} = 1 - \Phi\!\left(\frac{Q_n - 2T - \bar w}{\sigma_w}\right).

With the declared scatter only the far tail of the distribution is affected and QmaxQ_{max} is practically 1 (rejects of the order of 10−1010^{-10}); the share of loaves with T1 errors is 0.017 %, far below the 2.5 % allowed. A worse divider, a slower divider or a larger batch raise the scatter; once the T1 rule binds, the give-away grows linearly with the scatter and rejects start. In the bread line the T2 reject of the baking model contains the divider scatter together with the bake-loss scatter; when forming and baking are chained the check-weigher reject must be counted once (assumption a4). Crippled and double pieces and pieces torn by sticking dough are technology losses.

5. Unavoidable stops — maximum availability

Dough residues dry and ferment on pockets, knives, rounding tracks and moulding rollers, and the divider oil deteriorates; the equipment is cleaned dry and by hand once per production day (tclt_{cl}, declared 1 h). It is not cleaned in place, so the cleaning model does not apply. Over a production period tpert_{per}:

Amax=1−tcltper=0.958.A_{max} = 1 - \frac{t_{cl}}{t_{per}} = 0.958 .

6. Physics-bounded OEE

Scope of the bound. The theoretical label of OEEmaxOEE_{max} means physics-bounded under the declared scenario: the ceiling follows from the physics of the operation for the declared product, equipment and schedule, and it is a bound only conditional on those declarations. It follows the common definition of all models: AmaxA_{max} counts only stops required by regulation or forced by a physical mechanism, at their physical minimum duration, plus the start-up they force; PmaxP_{max} is the ratio of the physics-limited ideal cycle to the cycle of the capable process; QmaxQ_{max} is the share of conforming units, and mass losses of conforming product are reported only as yield.

OEEmax=Amax×Pmax×Qmax=0.958,Pmax=1.OEE_{max} = A_{max} \times P_{max} \times Q_{max} = 0.958, \qquad P_{max} = 1 .

The ceiling of this operation is set almost entirely by the daily cleaning; the legal weight rules act on the yield (give-away), not on the OEE, as long as the divider is capable. The point values and the Monte Carlo intervals over the declared uncertainty of every input are reported in the output table.

7. Assumptions, validity and use for other processes

See the assumption and validity lists generated from the configuration file. Most inputs of this model are declared scenario values, because no open source documents the weight scatter of dividers or the density change of dough in a hopper; the model shows how these quantities propagate through the legal rules and should be fed with the measured values of a line. The most influential inputs (sensitivity table) are the lot-mean uncertainty for the give-away, the divider scatter, the drift rate and the batch size for the reject share, and the cleaning time for availability.

Use for other processes. The codebook assigns forming to the bread process only. For other doughs and batters sold by weight (rolls, buns, cakes deposited into tins) the same construction applies as an approximation with their own nominal quantity, tolerable error, scatter and density drift; for products sold by count rather than weight the legal target does not apply and only the cleaning term remains.

References

  • Council Directive 76/211/EEC on the making-up by weight or by volume of certain prepackaged products, consolidated text of 11.04.2009.
  • International Organization of Legal Metrology. OIML R 87 Quantity of product in prepackages, Edition 2016.
  • U.S. Department of Energy. Bandwidth Study on Energy Use and Potential Energy Savings Opportunities in U.S. Food and Beverage Manufacturing, 2017.

7Parameters and provenance

Every input with its value, unit, role, uncertainty distribution and provenance. Open a quote to read the passage of the source that states the value.
  • QnQ_n800 g

    Nominal quantity of the prepackaged loaf (same as the baking model)

    Scenario · Fixed

    Declared assumption
  • TT15 g

    Tolerable negative error (tolerable deficiency) for a nominal quantity from 500 g to 1 000 g

    Regulatory · Fixed

    Verified quote
    Show verbatim quote
    from 500 to 1 000 — 15 from 1 000 to 10 000 1.5Council Directive 76/211/EEC of 20 January 1976 on the approximation of the laws of the Member States relating to the making-up by weight or by volume of certain prepackaged products (consolidated text 11.04.2009) (opens in a new tab). Publications Office of the European Union, 2009. p. 5. CELEX 01976L0211-20090411
  • kT2k_{T2}2

    Multiple of the tolerable negative error beyond which no prepackage may be marketed (T2 error)

    Regulatory · Fixed

    Verified quote
    Show verbatim quote
    no prepackage having a negative error greater than twice the tolerable negative error given in the table in 2.4 may bear the EEC signCouncil Directive 76/211/EEC of 20 January 1976 on the approximation of the laws of the Member States relating to the making-up by weight or by volume of certain prepackaged products (consolidated text of 11.04.2009) (opens in a new tab). Publications Office of the European Union, 2009. p. 5. CELEX 01976L0211-20090411
  • pT1p_{T1}2.5 %

    Maximum share of prepackages with a T1 error in a homogeneous group

    Regulatory · Fixed

    Verified quote
    Show verbatim quote
    A homogenous group of prepackages shall contain no more than 2.5 % of packages having T1 errors.OIML R 87 Quantity of product in prepackages, Edition 2016 (E) (opens in a new tab). International Organization of Legal Metrology (OIML), 2016. p. 13. OIML R 87:2016 (E)
  • Δavg\Delta_{avg}0 g

    Average requirement: the mean content may not be below the nominal quantity (offset zero)

    Regulatory · Fixed

    Verified quote
    Show verbatim quote
    The average actual quantity of product in prepackages shall be at least equal to the nominal quantity.OIML R 87 Quantity of product in prepackages, Edition 2016 (E) (opens in a new tab). International Organization of Legal Metrology (OIML), 2016. p. 13. OIML R 87:2016 (E)
  • zμz_{\mu}2.576

    Safety factor on the lot mean so that each lot meets the average requirement (one-sided 99.5 %), as in the filling model

    Scenario · Fixed

    Declared assumption

    Operating choice: Phi^-1(0.995), the same convention as the filling model. A larger factor raises give-away linearly.

  • σμ\sigma_{\mu}2 g

    Standard deviation of the lot mean of the dough-piece mass between weight checks (setting resolution of the pocket volume, density differences between mixer batches)

    Scenario · Uniform 1 to 3 g

    Declared assumption

    Declared assumption; replace with the drift of the controlled mean measured on the line.

  • σdiv\sigma_{div}6 g

    Standard deviation of the dough-piece mass from the divider at constant dough density (same declared value as the baking model)

    Technology · Uniform 3 to 10 g

    Declared assumption

    Capability assumption, identical to baking.sigma_divider so that the two models describe the same divider; replace with the divider's measured weight distribution.

  • rρr_\rho0.001 1/min

    Relative decrease of the dough density per minute in the hopper (gas production by the yeast)

    Scenario · Uniform 0.0005 to 0.002 1/min

    Declared assumption

    No open source with the density change of dough in a divider hopper was found. Declared assumption (0.05-0.2 % per minute, i.e. 1-4 % over 20 minutes); it depends on yeast level, dough temperature and the degassing by the divider ram.

  • MM300 kg

    Dough mass of one mixer batch divided from the hopper

    Scenario · Uniform 150 to 450 kg

    Declared assumption
  • npn_p4

    Number of pockets of the divider

    Scenario · Fixed

    Declared assumption
  • fsf_s25 1/min

    Strokes per minute of the divider

    Scenario · Uniform 15 to 35 1/min

    Declared assumption

    Installed rate of the divider; no open source with a physical limit of the stroke rate was found.

  • YY0.903226

    Bake yield, loaf mass per dough mass (output yield_max of the baking model)

    Property · Triangular 0.856255, 0.903226, 0.952311

    Computed

    Output `yield_max` of the baking model: 0.9032 dimensionless [0.8563, 0.9523].

  • tclt_{cl}1 h

    Daily cleaning of divider, rounder and moulder (dough residues, divider oil)

    Scenario · Uniform 0.5 to 1.5 h

    Declared assumption

    Dry and manual cleaning; the cleaning model (recirculated circuits) does not apply. Declared duration.

  • tpert_{per}24 h

    Production period of the availability balance (one production day)

    Scenario · Fixed

    Declared assumption

8Benchmarks

Documented values from open sources used to check the ceilings; no documented value may beat the theoretical limit.
BenchmarkCeilingValueProvenance
DOE bandwidth study: products without significant chemical reactions have a zero thermodynamic minimum (dough forming is not covered by the study; the principle applies)
Show verbatim quote
Most food and beverage products have a zero TM energy intensity because there are no significant chemical reactions involved.Bandwidth Study on Energy Use and Potential Energy Savings Opportunities in U.S. Food and Beverage Manufacturing (opens in a new tab). U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office, 2017. p. 25. DOE/EE-1571; OSTI ID 1512366
Theoretical0 kJ/kgVerified quote

9Unavoidable stops

Stops that physics or regulation impose; they set the maximum availability.
  • Daily cleaning of divider, rounder and moulderDeclared assumption

    Dough residues dry and ferment on pockets, knives, rounding tracks and moulding rollers, and the divider oil deteriorates; the equipment is cleaned dry and by hand once per production day.

10Assumptions

  1. The loaf weight is the dough-piece mass times the bake yield of the baking model; the scatter of the bake loss is counted in the baking model, not here. Within a batch the dough density falls linearly with time, so the piece mass drifts uniformly between the start and the end of the batch.
  2. The legal target follows the filling model: the mean must satisfy the average requirement with a safety factor on the lot-mean uncertainty and leave at most 2.5 % of loaves with T1 errors; dough pieces whose loaf would have a T2 error are removed at the check-weigher and reworked, and they count as quality losses; the give-away is reported as yield.
  3. No physical limit of the dough binds the rate: the stroke rate of the installed divider sets it, so P_max = 1 and the rate outputs carry no ceiling. Availability counts only the daily dry cleaning at a declared duration; the cleaning model does not apply.
  4. In the bread line the T2 reject of the baking model already contains the divider scatter together with the bake-loss scatter; when forming and baking are chained the check-weigher reject must be counted once. With the declared values both shares are below one in ten thousand.

11Range of validity

  • Volumetric dividers for wheat bread doughs sold as prepackaged loaves of 500 g to 1 000 g.
  • Approximation for depositing and moulding of other doughs and batters sold by weight, with their own scatter and density drift.

12Sensitivity

Elasticity: relative change of the output for a relative change of each parameter. Spearman rank correlation from the Monte Carlo draws is shown alongside.
  • tpert_{per}Production period of the availability balance (one production day)
    elasticity +0.044Spearman ρ —
  • tclt_{cl}Daily cleaning of divider, rounder and moulder (dough residues, divider oil)
    elasticity -0.043Spearman ρ -1.00
  • σdiv\sigma_{div}Standard deviation of the dough-piece mass from the divider at constant dough density (same declared value as the baking model)
    elasticity 0.000Spearman ρ -0.02
  • TTTolerable negative error (tolerable deficiency) for a nominal quantity from 500 g to 1 000 g
    elasticity 0.000Spearman ρ —
  • kT2k_{T2}Multiple of the tolerable negative error beyond which no prepackage may be marketed (T2 error)
    elasticity 0.000Spearman ρ —
  • YYBake yield, loaf mass per dough mass (output yield_max of the baking model)
    elasticity 0.000Spearman ρ 0.00
  • Output rises with the parameter
  • Output falls with the parameter

13Validation checks

A model is published only when every check passes.
  • Dimensional analysis17 outputs converted to their declared unitsPassed
  • Every cited value verified against its source5 verified, 0 failedPassed
  • No documented value beats the theoretical limitPassed
  • Ceilings in physical orderPassed
  • Fractions within 0 and 1Passed

14Causes of loss in this operation

Probability that each cause is documented as a relevant loss of a line, pooled over all segments.

15References

  1. Council Directive 76/211/EEC of 20 January 1976 on the approximation of the laws of the Member States relating to the making-up by weight or by volume of certain prepackaged products (consolidated text 11.04.2009) (opens in a new tab). Publications Office of the European Union, 2009. p. 5. CELEX 01976L0211-20090411
  2. Council Directive 76/211/EEC of 20 January 1976 on the approximation of the laws of the Member States relating to the making-up by weight or by volume of certain prepackaged products (consolidated text of 11.04.2009) (opens in a new tab). Publications Office of the European Union, 2009. p. 5. CELEX 01976L0211-20090411
  3. OIML R 87 Quantity of product in prepackages, Edition 2016 (E) (opens in a new tab). International Organization of Legal Metrology (OIML), 2016. p. 13. OIML R 87:2016 (E)
  4. Bandwidth Study on Energy Use and Potential Energy Savings Opportunities in U.S. Food and Beverage Manufacturing (opens in a new tab). U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing Office, 2017. p. 25. DOE/EE-1571; OSTI ID 1512366

Processes that use this operation