Sample size for determining in situ strength of concrete in structures

One of the most important tasks in the appraisal of an existing concrete structure is to establish the characteristic in situ strength for checking the structural adequacy of the structure for its intended usage. Existing codes of practice on the assessment of in situ concrete strength have addressed most of the important issues for carrying out this task, but one of the gaps is the appropriate size for the random sample. The approach taken by the codes at present is to leave this important decision to the judgement of engineers. This is not satisfactory. This paper presents a statistical study using a non-parametric method based on order statistics to establish the appropriate sample size of cores and non-destructive tests for determining in situ concrete strength. The statistical study reveals a relationship for the size of the random sample so that a desired fraction of the sample yields in situ strengths in a required range with a desired probability. An example is included to show the application of this relationship to determine sample size for the assessment of in situ strength for a multi-storey concrete building.