Multi-formulation and Production Planning in Feed Production: How Consolidated Formulation Transforms Planning

In feed mills with diversified portfolios, formulation and production planning typically operate with disconnected data. The formulator calculates each feed separately and delivers an ingredient list per product; production planning manually sums these consumptions to generate purchase orders. This flow works when the plant produces few SKUs with stable composition. When the portfolio grows, commodity prices fluctuate frequently, or inventory constraints affect multiple formulas simultaneously, the model begins generating costly and difficult-to-trace inefficiencies.

Multi-formulation solves exactly this problem. By optimizing all formulas of a production batch as a single mathematical problem, it produces a consolidated ingredient consumption that is fundamentally different from the sum of individual optimizations. This difference is not just technical: it changes what production planning can do, the precision with which it can do it, and the speed with which the company can react to raw material market changes.

Individual formulation versus multi-formulation: what changes in the resultIndividual formulation versus multi-formulation: what changes in the result

To understand the value of multi-formulation for production planning, it is necessary to understand why the sum of individual optimizations is not equivalent to a consolidated optimization.

When the formulator optimizes broiler starter feed independently, the algorithm finds the lowest-cost ingredient combination meeting that specific diet's nutritional requirements. The same is done for the grower feed, layer feed, and swine products. Each solution is locally optimal — the best possible considering only that product's constraints.

The problem is that individual formulations do not see each other. If corn has a high price, the algorithm will reduce corn in each formula to individually permitted minimums. But when summing all formulas for the production batch, one may discover that total soybean meal consumption exceeded available inventory, or that simultaneous inclusion of an alternative ingredient in all formulas created demand no prior quote covered. Each product was individually correct; the batch as a whole was not.

In multi-formulation, the mathematical problem is formulated differently. Nutritional constraints for each product are maintained, but the objective function minimizes total batch cost. Ingredient availability constraints are global: the available corn volume is a constraint applying to the sum of all formulas, not to each individually. This means the algorithm can distribute use of scarce ingredients across formulas in a coordinated way, slightly sacrificing one product's cost to make another's composition viable, resulting in total batch cost lower than the sum of individual optimizations could achieve.

The difference in consolidated consumption

Consolidated ingredient consumption is the central data that production planning uses to plan purchases, negotiate with suppliers, and feed the MRP. When this data comes from the sum of individual formulations, it represents the optimal consumption of each product in isolation, without considering formula interactions. When it comes from multi-formulation, it represents the optimal consumption of the batch as a whole, with all availability constraints and formula interdependencies already resolved by the optimizer.

In practice, this means that the consolidated consumption generated by multi-formulation is directly usable as the basis for MRP. Production planning does not need to make manual adjustments to handle ingredients whose total demand exceeds available inventory, because the optimizer already solved this problem within formulation. The purchase order generated by production planning from this data reflects what the plant will actually consume, not what it would consume in a hypothetical scenario where each product is produced independently.

How inventory constraints enter the optimization problem

One of the most directly relevant functionalities for production planning is the ability to insert ingredient availability constraints directly into the multi-formulation model. Instead of the formulator receiving a price list and calculating formulas without knowing what is in the warehouse, the model is fed with the available volumes of each ingredient and solves the entire batch considering these limitations as hard constraints.

Consider a concrete situation: the plant has 80 tons of canola meal in inventory, a product purchased as a price opportunity. Production planning wants to consume this inventory in the next production cycle. With conventional formulation, the nutritionist would have to manually adjust the minimum canola inclusion limits in each formula and run optimizations one by one, checking whether total consumption reaches 80 tons without exceeding limits that compromise each diet's quality. It is an iterative and time-consuming process.

In multi-formulation, the constraint is inserted once in the global model: total canola meal consumption must be equal to or greater than 80 tons. The optimizer distributes this ingredient across formulas where its inclusion has the least impact on cost and quality, respecting all individual nutritional limits. Production planning receives a coherent, executable consumption plan without manual iterations between formulation and planning.

Management of near-expiry ingredients

The same logic applies to near-expiry ingredients, a common situation in plants working with vitamin premixes, synthetic amino acids, and additives with relatively short use-by dates. Production planning must ensure these ingredients are consumed before expiry, but cannot simply double inclusions without checking the nutritional and economic impact on each diet. Multi-formulation allows defining a global minimum consumption constraint for the ingredient in question and letting the optimizer find the best distribution across batch formulas, ensuring utilization without compromising any product's technical specifications.

Production scheduling based on real formulation data

Sequencing production orders in a feed mill involves decisions that directly impact mixer setup time, high-turnover ingredient consumption, and the ability to meet delivery windows. The quality of these decisions depends on the quality of the data feeding them.

When production planning works with individually formulated consumptions, it faces a structural problem: data changes whenever the formulator updates a formula, but formulas are updated in a discoordinated manner over time. A formula may be reformulated due to an ingredient price change without the others being reviewed simultaneously. The consolidated consumption that production planning uses to plan purchases becomes outdated piecemeal, without a clear moment when all data is synchronized.

Multi-formulation solves this problem because batch optimization is a single, synchronized event. When the formulator runs the multi-product optimization for the next production cycle, all formulas are resolved simultaneously with current ingredient prices, currently available inventory, and production planning demands for that cycle. The resulting consolidated consumption is a coherent snapshot: all data references the same moment and the same set of constraints.

Grouping formulas by ingredient profile

The consolidated consumption generated by multi-formulation also allows production planning to make smarter decisions about production order sequencing. Formulas with similar ingredient profiles can be grouped to minimize ingredient changes between batches and reduce cleaning time between products sharing the same production line. This grouping is only possible when production planning has clear visibility into which ingredients participate in each formula and in what proportion, information that multi-formulation delivers naturally as part of its output.

In a plant producing poultry and swine diets on the same line, for example, production planning can identify that broiler starter and swine nursery formulas share a similar set of high-inclusion ingredients — corn, soybean meal, and vitamin premix — differing mainly in specific amino acids and additives. Producing these products in sequence before transitioning to formulas with very different compositions reduces the number of full setups, decreases cross-contamination risk, and increases shift productivity.

Integration between multi-formulation and MRP

MRP (Material Requirements Planning) works from two main inputs: demand for finished product and the technical structure of each product, which in the case of feed is the ingredient list and their proportions. When these inputs are accurate, MRP generates purchase orders and production orders that reflect what the plant actually needs. When inputs are outdated or inconsistent, MRP generates plans that need manual correction before being executed.

The biggest source of inaccuracy in MRP inputs at feed mills is the lag between formulation and planning. The formulator updates formulas in their software; production planning manually updates the technical structure in the ERP, or waits for someone to make that transfer. Between the formula update and the MRP update, there is an interval during which the production plan is based on incorrect data. If the substituted ingredient was corn for sorghum due to a price difference, the MRP will continue generating corn demand and no sorghum demand until someone corrects the record.

Formulation as a direct MRP feeder

Integration between formulation software and ERP eliminates this interval. When the formulator completes multi-product optimization for a production cycle, consolidated consumption is transmitted directly to the MRP as updated technical structure. Production planning starts receiving purchase orders generated based on the formulas that will actually be produced, not the formulas produced in the last cycle.

This integration has a particularly relevant effect when alternative ingredients participate variably in formulation. In plants working with DDGS, sunflower meal, brewery yeast, or other ingredients whose prices fluctuate frequently, these ingredients' inclusion changes from one production cycle to another depending on quotes. Without direct integration, each formulation change requires a manual ERP update — a process subject to errors and delays. With integration, the cycle closes automatically: new quote, new optimization, new consolidated consumption, new technical structure in MRP.

Supplier management and purchase negotiation based on consolidated consumption

Raw material procurement for feed involves ingredients with very distinct supply dynamics. Corn and soybean meal have active spot markets with daily quotes; vitamin premixes and synthetic amino acids are purchased from few suppliers with long lead times; alternative ingredients such as agroindustrial by-products have seasonal availability and case-by-case negotiated prices. Production planning must manage these dynamics in a coordinated way, and the quality of this management depends on the reliability of the consumption data it uses.

The consolidated consumption generated by multi-formulation allows production planning to work with more stable demand data for the planned production cycle. Instead of summing consumptions calculated at different times with different prices, multi-formulation consolidated consumption is calculated once with all ingredients priced simultaneously. The formulator and buyer work with the same number, referenced to the same market moment.

Sensitivity analysis as a procurement tool

Beyond consolidated consumption for the current cycle, parametric sensitivity analysis applied to multi-formulation provides production planning and procurement with information about how ingredient consumption changes in response to price variations. If corn prices rise 8%, what is the new corn consumption in each diet and in the total batch? At what point is substitution with sorghum or DDGS worthwhile considering the nutritional constraints of the complete portfolio?

These answers allow procurement to make proactive decisions. When sensitivity analysis indicates that corn has a price inflection point above which formulation migrates significantly to sorghum, the buyer can monitor corn prices relative to this reference value and prepare sorghum quotes in advance. The reaction window ceases to be formulation time plus quote time; it becomes only quote time, because formulation was already done in advance as scenario analysis.

Review cycles: synchronizing formulation and production planning

One of the operational challenges in formulation-production planning integration is defining the appropriate formula review frequency. Reviewing too frequently creates instability in procurement planning; reviewing too infrequently causes the plant to produce with formulas outdated relative to the ingredient market, missing cost reduction opportunities.

Multi-formulation facilitates review cycle definition because the optimization effort for an entire portfolio is practically the same as for an individual product. The formulator does not need to devote time proportional to the number of products; they configure the model once with all products, available ingredients, and nutritional constraints, update quotes, and run optimization. The time limiting review frequency becomes quote collection time, not formulation time.

In plants that adopt weekly multi-product review cycles, production planning receives updated consolidated consumption every week. Purchases with lead times over one week are planned in advance using data from the last review; short-term purchases are adjusted with data from the most recent review. This update rhythm would be unfeasible with individual formulation of each product separately.

Production planning as a strategic partner to the formulator

In many feed mills, the relationship between formulation and production planning is asymmetric: the formulator defines formulas and production planning executes planning based on them. Multi-formulation creates conditions for a more symmetric relationship, in which production planning contributes information that enters directly into the optimization model and influences the resulting formulas.

The inventory constraints that production planning inserts into the multi-formulation model are a clear example of this contribution. Production planning knows there is a specific volume of a given ingredient that must be consumed in the next cycle; the formulator knows to what extent that ingredient can be included in each diet without compromising nutritional specifications. Multi-formulation is the meeting point of these two perspectives: it finds the solution that meets production planning constraints within formulator-defined limits, without either needing to make concessions outside the model.

The same reasoning applies to supply disruption situations. If an ingredient is no longer available for the next production cycle, production planning can update the availability constraint to zero and re-run the multi-product optimization. The model recalculates all formulas simultaneously, redistributing available ingredients to meet portfolio nutritional requirements in the best possible way. The formulator validates the result and production planning receives a revised consumption plan in minutes, not hours or days.

Visibility of procurement decisions' impact on formulation

The integration between multi-formulation and production planning also creates visibility in the reverse direction: the formulator begins to see the impact of procurement decisions on formulation cost before purchases are made. If the buyer is evaluating closing a supply contract for a given ingredient at a fixed price for three months, the formulator can simulate in the multi-formulation model what happens to portfolio cost if that ingredient is locked at that price regardless of market fluctuations. This analysis informs the procurement decision with formulation data, creating an integrated planning cycle that most plants cannot yet execute due to lack of tools connecting these two areas.

Platforms like Formulamix allow formulators and production planning teams to work with the same optimization model, defining availability and minimum consumption constraints directly in the multi-formulation interface and generating consolidated consumption in structured format for ERP export. This type of operational integration is what transforms multi-formulation from a technical formulator resource into a strategic planning tool for the entire plant.

When multi-formulation generates more value

Multi-formulation generates more value in specific contexts that amplify the difference between global optimization and the sum of individual optimizations. Understanding these contexts helps production planning prioritize when implementing more frequent multi-product review cycles is worthwhile and when marginal gains justify the coordination effort.

Portfolios with shared ingredients across multiple diets are the most favorable context. The greater the ingredient overlap between formulas, the greater the chance the optimizer finds distributions that reduce total cost without compromising any individual product. A plant producing five formulas sharing eight main ingredients has much more to gain from multi-formulation than a plant producing five formulas with completely distinct compositions.

Production cycles with limited-availability ingredients also amplify the approach's value. When there are real ingredient volume constraints, the optimal distribution across formulas is a mathematical problem that human judgment has difficulty solving consistently. The optimizer handles these constraints systematically, ensuring the result is the best possible within existing limitations.

Portfolios with products of very distinct margins also benefit disproportionately more. In a portfolio where some products have very tight margins and others have higher cost tolerance, the multi-product optimizer can allocate high-quality, higher-cost ingredients to products where they make the most technical difference, and use more economical ingredients where nutritional requirements allow. This type of decision is very difficult to make consistently in individual product-by-product formulation.