Blogs
Summary
Tea manufacturers need to deliver a familiar taste using raw materials that change with every harvest. Connecting procurement planning and blending optimization helps teams account for changing raw-material characteristics, allocate scarce materials across products, and control sourcing and blending costs. Powered by ICRON’s AI-Native Decision Execution Hub, this approach combines optimization with AI-supported workflows to coordinate sourcing, allocation, and blend decisions.
The delivery arrives. The tasting changes the plan.
Imagine a tea blender starting the morning with samples from a newly arrived shipment. The origin is familiar. The grade is the one procurement ordered. The volume is enough for next week’s production.
But the cup tells a different story.
The tasting suggests that the new shipment may contribute less brightness than expected. Repeating the previous blend proportions may no longer deliver the brand’s familiar profile. The team needs to reassess the relevant material characteristics and available blending options.
Another tea in the warehouse could help. But it is already needed for a second blend, and its replacement is weeks away.
The blender can adjust today’s recipe. Procurement can look for more supply. Neither decision is complete until the team understands what happens to the rest of the plan.
How do you protect this blend without creating tomorrow’s shortage?
The same origin does not mean the same contribution
Growing conditions, harvest timing, processing, and storage affect tea characteristics. Differences in body, colour, brightness, and aroma can change which combinations are acceptable.
An established recipe is a starting point, not a guarantee that changing harvests, grades, or sourcing conditions will produce the same result. Teams need controlled flexibility: preserve required ingredients and inclusion limits while adjusting permitted proportions within approved quality boundaries.[1]
Illustrative proportions only, not a recommended recipe or evidence of sensory equivalence. Candidate blends must be evaluated against the relevant material characteristics, product requirements, and the manufacturer’s quality and sensory validation processes.
Fixing one blend can put another under pressure
The team identifies a suitable tea material that could help restore the desired profile. Allocating more of this material to the immediate blend may reduce the supply available for another product with fewer alternatives.
The decision must balance taste consistency, material allocation, and the cost of replenishment. It also needs to account for stock age and delivery timing. Having plenty of tea in total does not mean having the right combination available for every order.
Optimizing a single recipe can miss these dependencies. A low-cost blend today may leave the next run reliant on an expensive purchase.
When origin, grade and harvest keep moving: how sourcing, inventory and blend composition are planned as one decision.
Get the brochureThe buying decision starts with the future blend
Procurement receives an attractive offer. The price is competitive and the supplier can deliver. Should they buy?
That tea may need more of a costly complementary material to meet the target profile. A slightly more expensive material option might support several products and reduce pressure on scarce stock.
The value of a tea purchase depends on the blends it enables.
Evaluate the offer against expected demand, quality contribution, contracts, lead times, and existing inventory. A delayed shipment or a change in crop quality should trigger a fresh look at both sourcing and blending options.
Why this matters: FAO’s 2026 beverage-market analysis highlights exposure to extreme weather, supply constraints, and logistics disruption. A 2026 review in the Journal of Food Science and Technology also identifies the difficulty of managing consistency, cost, and blend selection through manual decisions alone.[4] [5]
ICRON Procurement Planning and Blending Optimization, powered by the AI-Native Decision Execution Hub
ICRON connects material characteristics, supplier options, purchasing costs, inventory, future demand, and blend requirements to evaluate sourcing and formulation decisions together across products, sites, and planning periods.
Procurement Planning helps evaluate what to buy, when, and in what quantities. Blending Optimization evaluates eligible combinations and proportions against defined product requirements.
Teams can compare supply scenarios, identify materials that restrict feasibility, and assess how a purchase changes the available blending options. The aim is to find a workable plan across demand, quality, and cost before committing stock or spend.[2]
Where agentic AI helps the team respond
Within configured workflows, requests or operational updates from planners, other agents, or connected enterprise systems can initiate a decision process. AI agents can gather relevant context, assess changes in demand, supply, material characteristics or costs, coordinate scenario evaluation, and initiate optimization runs. Within defined authority, they can carry out permitted planning actions or escalate decisions requiring approval.
AI agents
Assess requests and updates, coordinate optimization workflows, and execute authorized actions.
Optimization
Compare feasible choices against modeled quality, supply, and cost constraints.
Tea experts
Define the quality target, validate sensory performance, and approve critical changes.
This workflow depends on reliable data, configured integrations, and explicit rules. The tasting team’s expertise remains essential: a model needs validated quality inputs, and a proposed blend still needs the appropriate sensory and release checks.
A proven reference: Lipton Teas and Infusions
ICRON’s published Lipton customer story describes coordination involving 100 supply chain professionals in 15 countries across four continents. Reported results include:[3]
Customer-specific results from the procurement planning and blending optimization implementation. They are not a performance guarantee or results attributable solely to agentic AI.
For the team at the tasting table, the goal is practical: preserve the familiar cup while understanding what each adjustment means for tomorrow’s blends and commitments.
Practical questions for tea blending teams
How can we lower blend cost without changing the taste?
Compare eligible material combinations within defined quality requirements, including the cost and availability of complementary teas. Validate candidate blends through the established tasting process.
What if a shipment is delayed or an origin becomes unavailable?
Re-evaluate available material supply, incoming deliveries, transfers, and purchasing alternatives across affected products. If no feasible blend exists, make the constraint visible before revising commitments.
How can we reduce ageing stock without using scarce tea too early?
Consider stock age, release status, future demand, and alternative uses together. Allocate eligible lots across the portfolio instead of optimizing each recipe separately.
Can certification and customer-specific sourcing rules be respected?
Include applicable approved-source, certification, origin, and product-market restrictions as eligibility rules. The model needs current supporting data; optimization does not replace certification checks.
What data do we need, and does AI replace our tea blenders?
Start with the material characteristics used in planning, product quality requirements, available inventory, expected receipts, demand, supplier prices, and lead times. AI supports analysis and coordination; tea experts define and validate the sensory requirements.
Key Takeaways
- Plan with relevant material characteristics. Evaluate the quality contribution, availability, sourcing conditions, and cost of each material option rather than relying on origin or grade alone.
- Protect the portfolio. One recipe adjustment can consume material another product needs.
- Buy for blend feasibility. Evaluate price alongside quality contribution and future demand.
- Connect AI, optimization, and expertise. Coordinate decisions within approved rules and retain expert validation.
The next harvest will bring new characteristics.
How quickly can your team turn them into a feasible blending and procurement plan?
Related resources
Sources & further reading
- ICRON: What Is Tea Blending Optimization?
- ICRON: Procurement Planning and Blending Optimization for Tea & Coffee
- ICRON: Lipton’s Extraordinary Journey of Brewing Excellence
- FAO (2026). Price dynamics in global beverage markets: Trends, drivers and consequences.
- Abhiram, G. (2026). Mastering the art of tea blending. Journal of Food Science and Technology, 63, 836–855.
- Fomeni, F. D. (2018). A multi-objective optimization approach for the blending problem in the tea industry. International Journal of Production Economics, 205, 179–192.
- Moreira et al. (2024). Tea Quality: An Overview of the Analytical Methods and Sensory Analyses Used in the Most Recent Studies. Foods, 13(22), 3580.