The use of optimisation for enhancing the development of a novel sustainable masonry unit.

This paper examines whether it is possible to utilise optimisation techniques to enhance the efficiency of the experiment-based development of a new product. Conventionally, a laboratory-based experimental investigation requires to check all combinations of all independent parameters describing the product; hence the optimum levels of the tested parameters is obtained. As a result there may be many thousands of samples produced to optimise these parameters; this is a time consuming, costly and laborious process. Instead, it is suggested that an empirical model is established and validated using the results of the laboratory investigation making it suitable for optimisation. In this paper the empirical model describes the compressive strength of a new masonry unit (bitublock) in terms of two parameters that are the proportion of the coarse aggregate and the compaction pressure. The expression for the compressive strength obtained from all experimental data was optimised showing a similar result to that from an experimental investigation. A much reduced set of experimental data was then used to establish further empirical models. The predictions from these models were compared with the output from the full set of experimental data to assess the accuracy of these empirical models and arrive at recommendations for the reduction of an experimental programme for future studies. From the experimental investigation alone, the optimal compressive strength (35.82 MPa) of the new unit was obtained from a mix containing 30% coarse aggregate compacted at 24 MPa. Empirical modelling using only half of the original experimental data predicted that a coarse aggregate content of 31.95% and a compaction pressure of 21.73 MPa would provide an optimum compressive strength of 35.95 MPa. This shows that the use of optimisation techniques can improve the efficiency and economy of laboratory investigations and also has a potential to provide economies during future scaling-up and manufacturing processes, e.g. in this case, by reducing compaction pressures.