Industries / Construction and property / Commercial and residential construction / Concrete
Curing and testing
Curing keeps new concrete wet and within a safe temperature range long enough for the cement to react, and site testing checks each load and the hardening concrete against what was specified.
6
Milestones achieved, from c. 1920 to 2025.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
c. 1920
A cone for the site
The slump test fills a metal cone 12 inches (305 mm) high with fresh concrete in rodded layers, lifts the cone off and measures how far the concrete sinks; the wetter the mix, the more it slumps. A 1924 bachelor's thesis at Michigan Agricultural College says the test was "adopted by the Joint Committee in or about the year of 1920", first with a 6 by 12 inch cylinder and then with the 4 inch by 8 inch by 12 inch cone still in use, and that the Michigan State Highway Department had used it as its standard for three years. The same thesis found that "no two persons will puddle the concrete the same amount", so the result changed with the person doing it. ASTM C143 still describes the test.
LessonA test simple enough for any site gives comparable numbers only when every tester fills and rods the cone the same way.
Credit
the Joint Committee, as the thesis names it; critique by Chester I. Williams
1987
Strength from temperature history
Papers by Nurse (1949) and Saul (1951) in England, written about steam curing, gave the maturity rule: concrete of the same mix at the same maturity, counted as temperature multiplied by time, has about the same strength. When Skyline Plaza in Fairfax County, Virginia, collapsed during construction on 2 March 1973 and killed 14 workers (see Formwork), investigators at the US National Bureau of Standards (NBS, now NIST) found the slab where the failure likely began was four days old, had cured at about 7 °C, and that strength tables measured at constant temperature could not reliably tell how strong it was. The case renewed US interest in the maturity method, and in 1987 ASTM adopted C1074, a standard practice for estimating in-place strength from the concrete's logged temperature.
LessonStrength figures from a warm laboratory say little about a slab curing in the cold.
Credit
R. W. Nurse and A. G. A. Saul; Nicholas Carino and colleagues at NBS; ASTM
2005
Wet burlap within ten minutes
Bridge decks crack early when fresh concrete dries and shrinks. Specifications for low-cracking high-performance concrete (LC-HPC), written by the University of Kansas with the Kansas Department of Transportation, require one layer of saturated burlap on the deck within 10 minutes of strike-off, a second within 5 minutes, and a deck kept wet for 14 days. Contractors had to cast a trial slab first; for the first deck, LC-HPC-1, placed in October and November 2005, it took two attempts. A 2009 University of Kansas report on the first 14 LC-HPC decks found, in early crack surveys, that they were performing better than matched control decks.
LessonThe report found "clear and consistent communication between the contractor, owner, and testing personnel" vital, because a rule timed in minutes works only if someone on the deck times it.
Credit
David Darwin, JoAnn Browning and Heather McLeod, University of Kansas; Kansas DOT; Transportation Pooled Fund study TPF-5(051)
2015
Bubble spacing measured fresh
Concrete that freezes while wet needs many tiny air bubbles, close together, for the water to expand into. The standard pressure meter (ASTM C231) measures only how much air there is. The Super Air Meter (SAM), developed at Oklahoma State University, runs the same test at a series of higher pressures and in just over 10 minutes gives a "SAM number" for the quality of the bubble system. By the end of 2015 a pooled-fund study led by the Oklahoma DOT had tested it on more than 100 laboratory mixtures, 50 field mixtures and more than 30 mixtures run by the FHWA Turner-Fairbanks laboratory, and a SAM number of 0.20 matched the 0.008 in (0.2 mm) bubble spacing measured on hardened samples under a microscope (ASTM C457) 94% of the time.
Credit
Tyler Ley and his team at Oklahoma State University; Oklahoma DOT; FHWA
2020
Pour temperatures predicted
A thick pour such as a bridge footing heats up as the cement reacts, and if its core gets much hotter than its surface, the surface cracks. ConcreteWorks, software developed at the University of Texas at Austin for the Texas DOT, predicts those temperatures from the mix, the shape and the weather. For the Iowa DOT, Iowa State University adapted it to Iowa materials and weather and checked it on a monitored bridge footing: the adapted version predicted a peak of 149.16 °F (65.1 °C) at 52 hours, against 149.00 °F measured at 50 hours.
LessonIowa found that the program's default materials, its four Iowa weather stations and its 14-day output limit did not fit local practice, so a prediction tool has to be fitted to local materials before it is trusted.
Credit
researchers at the University of Texas at Austin (TxDOT project 0-4563); Kejin Wang, Iowa State University
2025
Enough compound beats even compound
State highway agencies widely cure concrete pavement by spraying a membrane-forming curing compound (ASTM C309) that slows evaporation. Some US states accept the coat by comparing it with a white sheet of paper. For the Wisconsin DOT, researchers watched curing on four paving jobs in 2023 and 2024 and tested panels in the laboratory. They found field application "highly variable", with curing carts running at 50 to 100 ft (15 to 30 m) a minute in short bursts. In most weather a sufficient but uneven coat protected the concrete better than an even coat that was too thin, and compound sprayed while bleed water still sat on the surface let more water escape.
LessonHow much compound goes on matters more than how even it looks, which an eye check cannot judge.
Credit
John Kevern, University of Missouri–Kansas City; Danny Xiao, University of Wisconsin–Platteville; Wisconsin DOT (project 0092-22-03)
Watch for
Radar checks the curing coat
The Wisconsin team says the surface dielectric constant measured with ground-penetrating radar (GPR), which rises with surface moisture, can map poorly cured areas and confirm complete coverage, and that an "Evaluation Index" linking laboratory evaporation results to radar readings sets the right application rate for the forecast weather.
Achieved whena team outside the project reproduces the radar coverage maps against weighed application rates on a paving job, or a state other than Wisconsin writes the check into a specification.
Claimed by
John Kevern (now at the National Laboratory of the Rockies), Danny Xiao and Dan Zollinger, at the National Concrete Consortium, April 2026
Open
01
Score mass pour temperature forecasts
A public set of measured temperature histories from at least 20 instrumented thick pours, such as footings, pier caps and building mats, each with its mix, dimensions, placing temperature, insulation, cooling and weather, gathered from published transportation department reports and from contractors willing to share sensor logs. Then score ConcreteWorks, a simple chart method, and a small machine-learning model against the measurements.
The briefCredit · unclaimed02
Curing coverage from a phone
A phone app that photographs white curing compound on fresh concrete beside a printed gray reference card and estimates the application rate in ft² per gallon (m² per liter). Train it on small mortar panels sprayed at weighed rates and photographed in sun, shade and overcast light, then check it against the tank-volume method on paving jobs, with photos taken by the agency's inspector from outside the work zone.
The briefCredit · unclaimed
Open 01 · the brief
Score mass pour temperature forecasts
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public set of measured temperature histories from at least 20 instrumented thick pours, such as footings, pier caps and building mats, each with its mix, dimensions, placing temperature, insulation, cooling and weather, gathered from published transportation department reports and from contractors willing to share sensor logs. Then score ConcreteWorks, a simple chart method, and a small machine-learning model against the measurements.
Why it is still open
Published checks test one tool on one structure. Iowa State compared ConcreteWorks and 4C-Temp&Stress on one footing, and a 2022 Georgia DOT study built a nomogram from one bridge site. The only open temperature data set found, released in 2025 by Cavadia and colleagues at the Universidad Industrial de Santander, covers a 1.2 m slab-on-beam model, and its authors write that "few experimental outdoor data on the internal and external temperatures of concrete structures are available". Several mass concrete calculators appeared on GitHub in 2026, but none holds measured pours.
Built with what exists
Published reports such as Iowa DOT TR-712 and the Georgia DOT Phase II study; ConcreteWorks; Python with pandas and scikit-learn; an AI model that reads tables and plots in PDF reports to pull out logged temperatures, with every extracted value checked by a person.
First step
Put the Iowa footing and Georgia case-study records into one CSV format and run ConcreteWorks on both.
Done looks like
At least 20 pours, each with at least three sensor positions, released under an open license, and a table of each method's error on peak temperature and on the largest core-to-surface difference, in °C, and on time to peak, in hours.
Open 02 · the brief
Curing coverage from a phone
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A phone app that photographs white curing compound on fresh concrete beside a printed gray reference card and estimates the application rate in ft² per gallon (m² per liter). Train it on small mortar panels sprayed at weighed rates and photographed in sun, shade and overcast light, then check it against the tank-volume method on paving jobs, with photos taken by the agency's inspector from outside the work zone.
Why it is still open
The Wisconsin DOT's 2025 brief says "there is no existing method for quantifying or systematically evaluating the quality or effectiveness of curing in the field." Its researchers report that California's pad test (CT535), which weighs pads laid under the sprayer, is "not a practical test for field QC/QA" and that pad and tank figures "vary a lot". They show image-based coverage percentages on laboratory panels, and their field answer is ground-penetrating radar (see Watch for), which needs a radar unit and a trained operator. Earlier work stops short of a rate. FHWA found in 2006 that a portable paint reflectometer improved estimates of how much white compound went on, but it is a dedicated instrument. Soon Ho Baek, Kang In Lee and Seong-Min Kim of Kyung Hee University (Sensors, 2023) mounted a camera on the spraying equipment and judged each area sprayed or not by a grayscale cutoff (185 in sun, 170 under cloud), while a flowmeter measured the amount.
wisconsindot.govcptechcenter.orgfhwa.dot.govpmc.ncbi.nlm.nih.gov
Built with what exists
A phone camera; a printed gray card; OpenCV; Meta's SAM 2 model to outline the slab in each photo; 300 by 300 mm mortar panels, a hand pump sprayer and a 0.1 g scale; Python for the fit.
First step
Spray 20 panels at four rates, weigh them, photograph them in sun and in shade, and see whether brightness tracks the rate.
Done looks like
On 100 laboratory panels at five known rates, the app's estimate within 15% of the weighed rate on at least 80%; then the same check on 20 paving sections against tank-volume figures, with all photos and weights published.