Weight Comprehensive weighted regression models and their statistic indices Optimized process parameters DEC, λ1 PRR, λ2 Comprehensive weighted regression model Coefficient of determination, R2 Lack-of-fit test statistic, P Heat pump drying temperature [°C] Moisture content at transition point of drying stages [%] Microwave drying power [W] 0.1 0.9 Y = 85.03 - 12.84X2 - 14.74X3 - 28.21X12 - 17.87X32 0.8811 0.4284 34.44 33.02 202.73 0.2 0.8 Y = 85.20 - 10.83X2 - 13.31X3 - 27.08X12- 13.95X22 - 16.68X32 0.8940 0.4665 34.88 36.34 202.52 0.3 0.7 Y = 85.37 - 11.88X3 - 25.95X12 - 18.11X22 - 15.50X32 0.9085 0.5059 35.07 37.92 203.71 0.4 0.6 Y = 85.54 - 10.45X3 - 24.82X12 - 22.28X22 - 14.32X32 0.9226 0.5359 35.16 38.82 205.47 0.5 0.5 Y = 85.72 - 8.99X3 - 23.67X12 - 26.50X22- 13.07X32 0.9342 0.5375 35.21 39.42 207.38 0.6 0.4 Y = 85.88 - 7.59X3 - 22.58X12 - 30.61X22 - 11.95X32 0.9413 0.4967 35.24 39.83 210.10 0.7 0.3 Y = 86.05 - 21.42X12 - 34.77X22 - 10.77X32 0.9439 0.4208 35.25 40.14 213.43 0.8 0.2 Y = 86.21 - 13.19X1X2- 20.29X12 - 38.94X22 - 9.59X32 0.9429 0.3398 35.25 40.37 217.30 0.9 0.1 Y = 86.38 - 16.64X1X2 - 19.16X12 - 43.10X22 0.9394 0.2759 35.25 40.53 222.89

Table 5: Regression models and their optimized process parameters when setting DEC and PRR at variable weights.