PSI - Issue 5

Adelaide Cerveira et al. / Procedia Structural Integrity 5 (2017) 1116–1122 Author name / Structural Integrity Procedia 00 (2017) 000 – 000

1118

3

which effects are significant. A two-way ANOVA assumes that the observations within each cell are normally distributed and assumes the homogeneity of variances, McCullagh and Nelder (1989). All of these statistical analyses were conducted using SPSS (IBM SPSS Statistics 22), Marôco (2007). A two-way ANOVA with interaction was conducted to evaluate the effects of the humidity and the percentage of ashes in the mixture on compressive strength. The humidity factor includes three levels: 25%, 50% and 75%. The factor filler/precursor ratio, corresponding to mine tailings by fly ash ratio, includes five levels: 80/20, 60/40, 40/60, 20/80 and 0/100. In order to simplify the notation, these five cases are solely denoted by 20%, 40%, 60%, 80%, 100%, corresponding to the percentage of ashes in the mixture. The interaction factor include fifteen combinations: 25%&20% (humidity=25% and ashes percentage=20%), 25%&40%, 25%&60%, 25%&80%, 25%&100%, 50%&40%, 50%&60%, 50%&80%, 50%&100%, 75%&40%, 75%&60%, 75%&80%, 75%&100%. In figures 1, 2 and 3 are presented boxplots to show the distributions of compressive strength for each of the factors and interaction.

Fig.1. Boxplots of compressive strength for different percentage of ashes

Figure 1, shows that the medians of the five groups are different and, in general, increase as the percentage of ash increases. Moreover, there are also differences in terms of variability of the data, the filler/precursor ratio that corresponds to smaller variability in compressive strength is the case 60/40, corresponding to 40% of ashes. Greater values of compressive strength are obtained in the mixture with 100% of ashes.