Prediktor Kekuatan Bayesian

What is characteristic strength?

Characteristic strength is the value of compressive strength below which no more than 5% of test results are expected to fall. In structural design codes such as Eurocode 2 (EN 1992) and IS 456, the characteristic strength (f_ck) is the primary parameter used to specify concrete grades — for example, C30 means the characteristic strength is 30 MPa. The concept exists because concrete is inherently variable: even under controlled factory conditions, the strength of individual test specimens will scatter around a mean value. Rather than using the mean strength (which 50% of batches will fall below), engineers use the 5th-percentile value to ensure that the vast majority of the concrete placed on site meets or exceeds the design requirement.

Mathematically, the characteristic strength is typically computed as f_ck = f_cm − 1.645σ, where f_cmis the mean compressive strength and σ is the standard deviation. The factor 1.645 corresponds to the 5th percentile of a normal distribution. In practice, the actual distribution of concrete strength is not perfectly normal — it tends to be slightly right-skewed — but the normal approximation is widely accepted for design purposes. Our Monte Carlo approach avoids this assumption entirely by computing the characteristic strength directly from the simulated sample distribution.

Why does concrete strength vary?

Concrete strength variability arises from multiple sources, each contributing uncertainty to the final compressive strength. The water-to-cement (w/c) ratio is the most influential factor: even a small variation in the amount of mixing water — due to aggregate moisture content, batching accuracy, or water added on site for workability — can shift the w/c ratio by 0.02 to 0.05, which translates to a strength change of 3 to 8 MPa. Cement itself varies from batch to batch; the fineness, mineral composition, and alkali content all affect the rate and extent of hydration.

Aggregate properties contribute additional scatter. The shape, surface texture, and mineralogy of aggregate particles influence the bond between aggregate and cement paste. Crushed aggregates with angular surfaces generally produce stronger concrete than rounded gravel, but the degree of improvement depends on the specific rock type. Aggregate grading — the distribution of particle sizes — affects the packing density and therefore the volume of cement paste required to fill the voids. Poorly graded aggregate increases paste demand and reduces achievable strength.

Construction practices introduce further variability. Mixing time and intensity, transport duration, placing and compaction methods, and the quality of curing all affect the final strength. Concrete that is inadequately compacted may contain voids that reduce strength by 5 to 10% per 1% of entrapped air. Similarly, concrete that dries out prematurely during curing can lose 15 to 30% of its potential 28-day strength because hydration ceases once the internal relative humidity drops below about 80%.

How does curing temperature affect strength?

Temperature is a critical factor in concrete strength development, governed by the Arrhenius equation for the rate of chemical reactions. At higher temperatures, the hydration of cement proceeds faster: concrete cured at 35°C may reach 70% of its 28-day strength within 7 days, while the same mix cured at 10°C might only reach 40% at the same age. However, this early-age acceleration comes at a cost. High curing temperatures produce a less uniform microstructure with coarser pores, which can reduce the ultimate (long-term) strength by 10 to 15% compared to concrete cured at moderate temperatures around 20°C.

The Nurse-Saul maturity method, used in our predictor, quantifies this relationship by computing a maturity index that combines time and temperature into a single parameter. The maturity is the integral of temperature above a datum (typically −10°C) over the curing period. Two concrete specimens with the same maturity index will have approximately the same strength, regardless of their individual time-temperature histories. This principle is widely used in construction to estimate in-place strength for formwork stripping decisions and to optimise curing schedules for precast elements.

Cold weather concreting (below 5°C) presents particular challenges. If fresh concrete freezes before reaching a compressive strength of about 3.5 MPa (500 psi), the expansion of freezing water in the capillary pores can permanently damage the microstructure, reducing ultimate strength by 20 to 40%. For this reason, most specifications require protective measures — insulated formwork, heated enclosures, or chemical accelerators — when ambient temperatures fall below 5°C. Conversely, in hot climates, the use of chilled mixing water, ice, or cooled aggregates helps keep the concrete temperature below 30 to 35°C at placement to avoid thermal cracking and long-term strength loss.

How our predictor works

This tool runs a Monte Carlo simulation with 5,000 samples directly in your browser using a Web Worker. For each sample, the simulator draws random values for the w/c ratio, curing temperature, and material quality from distributions centred on your inputs. It then computes the compressive strength for each sample using Abrams’ law (with Bolomey corrections for different cement types) and the Nurse-Saul maturity factor. The result is a histogram showing the full distribution of likely strengths, along with key statistics: mean, median, 5th and 95th percentiles, and characteristic strength.

You can also enter a target strength to see the probability that a random batch will exceed it. This is particularly useful for specifying concrete: if you need a characteristic strength of 30 MPa, you can experiment with different w/c ratios and curing conditions to find the combination that gives you a comfortable margin above your target. All calculations run locally in your browser — no data is sent to any server, and the simulation completes in under a second on modern hardware.