# AI learns from your concrete strength tests and proposes leaner mixes, so ready-mix plants use less cement without missing strength (Tomlinson Ready Mix, Trio Ready-Mix, VCNA Prairie Materials)

> AI proposes leaner concrete mixes from your test data: Tomlinson cut 35 kg cement per m3 (8.8%) in 6 weeks; Trio optimized 34% of production at $3/m3 less.

- Canonical URL: https://aiagentsformanufacturing.com/use-cases/ai-ready-mix-concrete-mix-design-optimization-less-cement-tomlinson-trio-prairie/
- Department: Quality (https://aiagentsformanufacturing.com/departments/quality/)
- Evidence level: Verified. Real deployment at a named or precisely described company, with a measured result from a primary source.
- Added: 2026-11-17
- Last updated: 2026-11-17
- Publisher: AI Agents for Manufacturing (https://aiagentsformanufacturing.com)
- Manufacturers named: Tomlinson Ready Mix (https://aiagentsformanufacturing.com/companies/#tomlinson-ready-mix), Trio Ready-Mix (https://aiagentsformanufacturing.com/companies/#trio-ready-mix), VCNA Prairie Materials (https://aiagentsformanufacturing.com/companies/#vcna-prairie-materials), Heidelberg Materials (https://aiagentsformanufacturing.com/companies/#heidelberg-materials)
- AI tools deployed: Giatec SmartMix (https://aiagentsformanufacturing.com/tools/#giatec-smartmix), Concrete.ai Concrete Copilot (https://aiagentsformanufacturing.com/tools/#concrete-ai-concrete-copilot)

## Role

Quality control and technical services in concrete and building materials plants (mix design, strength testing). Typical users: quality control manager, technical services manager, ready-mix plant manager, general manager of a concrete or precast producer, sustainability lead

## The problem

Cement is the most expensive and most carbon-intensive ingredient in concrete, and because a failed strength test is costly, producers tend to over-design their mixes with extra cement. The test data that could show where the margin is sits in folders and spreadsheets, and analysing it by hand takes time that quality teams do not have. Tomlinson Ready Mix (Ottawa, Ontario, Canada) managed mix designs and test data with physical folders and customised spreadsheets that could not keep pace as it grew, so the team relied on intuition and experience to make mix design decisions. On one project where the client required 100% of design strength in 7 days, Tomlinson deliberately over-designed the mix. Trio Ready-Mix (Southern Vancouver Island, British Columbia, Canada, part of the Ralmax Group) managed its high-volume mixes manually across legacy systems that did not talk to each other, so gathering and analysing data limited real-time decisions; its general manager said a task that took at least 20 minutes by hand is done by the software in a tenth of a second.

## How AI helps

The mix management software stores each plant's mix designs, batch records and strength test results, and a machine learning model learns how each mix actually performs with the plant's own materials. It then flags mixes with more strength margin than needed and proposes small, incremental changes (less cement, different proportions) that the quality team approves, while strength results keep coming in to confirm them. At Tomlinson and Trio, Giatec SmartMix with its AI, called Roxi, compared historical performance data with current mixes; at Tomlinson the data showed the 7-day strength target was being reached in 3 days, so there was room to optimize. Trio found the general model needed a period of adaptation to its plant, made a conservative first adjustment, collected more data and then made a more aggressive one. At VCNA Prairie Materials (Illinois, USA), Concrete.ai's Concrete Copilot was integrated with the producer's existing software and generated mix designs from its own data and local materials; the platform creates millions of candidate mixes, predicts their hardened performance and presents the best one for the producer to approve or modify.

## Step by step

1. Gather one or two years of mix designs, batch records and strength results (7, 28 and 56 day) with the specified strength for each mix.
2. Calculate for your highest-volume mixes how far the average strength sits above the required strength and how much the results vary.
3. Pick two or three high-volume mixes with the most margin and agree with your engineer what minimum margin you must keep under your standard (CSA, ACI or local).
4. Make small changes one at a time (about 2% of cement, as Tomlinson did), with extra testing on the first loads, before the next step.
5. Keep the testing frequency up after each change and set an alarm if the results drift toward the minimum.
6. Track cement per cubic metre, cost per cubic metre, CO2 per cubic metre and the share of production on optimized mixes every month.

## Tools

- Giatec SmartMix with the Roxi AI: used at Tomlinson Ready Mix and Trio Ready-Mix
- Concrete.ai Concrete Copilot: used at VCNA Prairie Materials
- Other tools to evaluate: the mix optimization features of your batching or quality control software, AI optimization tools from other concrete technology suppliers, statistical analysis of your own strength data
- To start without new software: your mix designs and strength test results and any LLM (ChatGPT, Claude, Copilot, Gemini) with the prompt below

## Result

Tomlinson Ready Mix (Giatec case study, the vendor): three incremental adjustments of about 2% of cement each cut a combined 35 kg of cement per cubic metre in 6 weeks with no trial batches, an 8.8% cement reduction and a 5% cost reduction on that mix, without compromising performance or quality. Trio Ready-Mix (Giatec case study): cut cement by 50 kg per batch; it took two months to see tangible benefits, and by the 10th month 34% of overall production had been optimized, with an average reduction of $3 per cubic metre across the optimized mixes. VCNA Prairie Materials (Concrete.ai announcement, the vendor): significant reductions in cost and carbon footprint, no figure for the plant alone; across Concrete.ai's field testing with US producers, mixes used in over 2 million cubic yards of concrete were optimized, with average material savings of $5.04 per yard and an average carbon reduction of 30%. Heidelberg Materials, an investor in Giatec, says Giatec's AI-driven optimization tools help it reduce cement use by up to 20% in its concrete mixes.

As reported by the source; most figures are vendor or self-reported and not independently verified.

## Human review: cautions

- All figures come from the software vendors (Giatec, Concrete.ai) or from Heidelberg Materials, which invested in Giatec; they are not independently audited.
- Tomlinson's 8.8% and 5% apply to the mix of one project, not to its whole production; Trio's savings apply to the 34% of production that was optimized.
- Concrete.ai's $5.04 per yard and 30% carbon reduction are averages across its field testing with many producers, not VCNA Prairie Materials' own result.
- Mix changes remain the responsibility of your qualified engineer and must meet your standard and the project specifications; keep testing up after every change.
- The model needs enough of your own test data per mix; Trio needed about two months before the suggestions paid off.

## Try it yourself (prompt)

```text
You are a concrete quality control engineer. I want to find which of our mixes are over-designed and how much cement we could safely take out.

I will paste: (1) our mix designs [mix code, specified strength, cement and supplementary materials in kg per cubic metre, water, admixtures, volume produced last year]; (2) our strength test results [mix code, date, age in days, strength result]; (3) our cement cost per tonne and the standard we follow (CSA, ACI or other).

Do this:
1. For each mix, calculate the average strength, standard deviation and margin over the specified strength at each age, and list the mixes with the largest margin and volume.
2. Using the rules of our standard for required average strength, estimate how much margin each top mix really needs, and state the formula and assumptions you used.
3. For the top three mixes, propose a step-by-step plan of small cement reductions with the extra tests to run after each step.
4. Estimate the cement saved per year and the cost saved at our cement price, as a range.
5. List the risks (material changes, weather, early-age requirements) and what would make us stop or reverse a change.

The mix changes must be approved by our qualified engineer before use. Use only the data I paste in; if something is missing, write 'not found' or list it as a question instead of guessing. Do not send, order, post or change anything in any system: produce drafts for my review and approval.
```

## Related use cases

- [Deep learning optical sorters spot hard defects in fruit and nuts on the line, so food plants pack cleaner product with fewer hand sorters (Chelan Fruit, Kings River Packing, The Almond Company)](https://aiagentsformanufacturing.com/use-cases/ai-deep-learning-optical-sorting-grading-food-defects-chelan-kings-river-almond-company/)
- [AI double-checks what the AOI machine flags on circuit boards and clears the false calls, so operators only review real defects (Siemens Rastatt, USI, FermionX)](https://aiagentsformanufacturing.com/use-cases/ai-aoi-false-call-reduction-smt-pcb-inspection-siemens-rastatt-usi-fermionx/)
- [AI reads the date code and checks the label on every pack at line speed, so the wrong label or code never ships (Ilchester Cheese, Del Monte Foods)](https://aiagentsformanufacturing.com/use-cases/ai-label-date-code-verification-ocr-packaging-line-ilchester-del-monte/)
- [AI predicts which boards really need the slow test, so X-ray and end-of-line test stations stop being the bottleneck (Siemens Amberg, ZF, Flex)](https://aiagentsformanufacturing.com/use-cases/ai-predictive-test-selection-skip-xray-end-of-line-tests-siemens-zf-flex/)

## Sources

- https://www.giatecscientific.com/case-study/tomlinson-ready-mix-giatec-smartmix/
- https://www.giatecscientific.com/case-study/trio-ready-mix-giatec-smartmix/
- https://www.concrete.ai/news/concrete-ai-applies-generative-ai-solution-to-carbon-wise-mix-designs/
- https://www.heidelbergmaterials.com/en/digital/lets-build-the-future-together

---
Please cite the canonical URL: https://aiagentsformanufacturing.com/use-cases/ai-ready-mix-concrete-mix-design-optimization-less-cement-tomlinson-trio-prairie/
