SITELIVE 
What was reported
Engineering News-Record published an account on 14 August 2026 of GreenVibe, a Ramat Gan company selling sensors that are cable-tied to reinforcement before a pour and stay inside the element for the life of the structure. The company has raised US$12 million to date, says it only recently moved to general commercial availability, and says roughly 70 to 80 per cent of the approximately 10,000 sensors it has produced are on active projects, with the rest used for model training and validation. Chief executive and co-founder Roee Reshef says there are 11 paying projects, with the contractors Solel Boneh, BST and Tidhar each developing further pipelines, and that a pilot has been completed with a large US contractor that cannot be named on the record.
The story that gives the technology its weight is a disagreement, not a demonstration. On the Beyond office tower in Givatayim, immediately east of Tel Aviv — roughly 1,050 ft, about 320 metres — the embedded sensors indicated the concrete had not developed enough strength for the next operation while conventional laboratory testing indicated the work could proceed. Project manager Shalom Buchnik told ENR the team did not fully follow the sensor data, and the resulting problem required more than US$500,000 in repair and replacement. When the two sources disagreed again on a critical element, the team waited: post-tensioning was delayed until 11 days after placement, and Reshef says subsequent core testing supported the sensor reading. Buchnik says the system also helped the project save 98 working days overall.
Why the measurement is different from maturity
Most in-place strength estimation on site today is the maturity method. ASTM C1074 requires you to establish a strength–maturity relationship for that specific mixture in the laboratory, then record the temperature history of the element and read strength off the curve; the practice is written for exactly the decisions that hold up a floor cycle — formwork removal and reshoring, post-tensioning of tendons, ending cold-weather protection, opening pavement to traffic. Its own limitations clause is blunt: the method does not account for the effect of early-age temperature on long-term strength, and it needs to be supplemented by other indications of the potential strength of the field concrete.
GreenVibe’s pitch is that it measures the material rather than inferring from its thermal history. The company says each sensor combines several sensing methods — its product overview lists five technologies, and ENR names ultrasound, flexoelectric measurements, electrical resistivity and temperature — in an IP67 enclosure, feeding a hub and a cloud platform where machine-learning models turn the raw signals into a strength curve on a phone. Reshef says combining those channels lets the system derive strength from physical change in the material without first building a mix-specific calibration curve, which is what would let one system follow many mixes and curing conditions on one job. That single claim is the whole product, and it is the claim with the least independent evidence behind it.
The honest limits
It does not replace the acceptance test. GreenVibe is not approved to substitute for required concrete acceptance testing in the US, and Reshef puts it as an additional quality-assurance layer that does not replace 28-day testing. The same holds here: compliance of supplied concrete in Australia runs through the sampling and testing regime of AS 1379, on standard-cured specimens, and nothing embedded in a slab changes what the supplier and the specification are assessed against. The company is participating in ASTM Committee C09 and Subcommittee C09.64, which owns C1074 — and a revision of C1074 is under development as work item WK97581 — but participation in a committee is not an endorsement by one.
Coverage is local, not elemental. Reshef estimates the effective measurement radius at roughly 8 in., about 200 mm, depending on wave behaviour, so a honeycomb or a cold joint a metre away from the nearest sensor is simply not in the data. And every performance number in circulation is a party’s own: the 98 saved working days on Beyond and the US$500,000 repair are the project manager’s account; the 200 sensors across 80 placements at Ashtrom’s Herzliya project from June 2025, the C40 concrete reaching about 70 per cent of target strength in an average of 72 hours, the roughly 40 days saved, and the approved grade and cement-content reductions at Tidhar projects are the company’s figures as reported; the “92% data accuracy” on GreenVibe’s own site carries no published methodology, no test population and no comparison against cores. There is no independent validation study in the public record, and the deployment base is one market plus a pilot.
What it means for operators
The transferable lesson is not about sensors. It is that the moment you have two sources of truth on a strength-governed hold point, you need a written rule about which one governs before the pour, not an argument on the deck at 6am. Decide it per decision — stripping, back-prop removal, stressing, loading — and put it in the ITP: which evidence, what threshold, who signs, and what happens when the two readings diverge. The expensive half of the Beyond story was not a bad instrument, it was an unresolved decision rule.
Then treat instrument placement as engineering. A sensor with a 200 mm radius belongs in the governing location — the thickest section, the coldest exposed face, the congested zone the vibrator struggles to reach — not the one closest to the access ladder. Anything that leads to a designer-approved reduction in grade or cement content has to carry that approval in the project record, and the raw curves need to survive longer than the app subscription: if this data ever backs a claim or a defect argument, it will be read years later by someone who was not there.