The research team then proceeded to conduct linear correlation analysis to determine the
correlation between fuel/cycle and cycle time components and payload. Table 6 and
Figures 6-9 show correlation coefficients with their corresponding p-values and scatter
plots, respectively. Surprisingly, there was no statistically significant correlation between
payloads for the experimental period and fuel/cycle as indicated by the p-value of 0.1801
(greater than α = 0.05). This was contrary to expectation and hence the correlation
between payload for the entire available data set (May 3 to July2) and fuel/cycle was also
analyzed. This yielded a statistically significant correlation (p-value of 0.0000). Modeling
fuel/cycle per ton is desirable so that the model can be extended to different truck
payloads. In fact, it is expected that fuel consumption should correlate to amount of
material carried since more work is done. Hence, correlations between cycle time
components in Table 6 and fuel/cycle/ton was tested and statistically significant positive
correlation was found.