Figure 5
A line graph compares the average computation time in seconds for two planners, OPTIC and POPF, across various disrupted-operation instance scales. The y-axis represents the average computation time in seconds, ranging from 10ˆ0 to 10ˆ3. The x-axis lists different instance scales: S1, S2, M1, M2, L1, L2, L3, VL1, VL2, and VL3. The graph includes two data lines: one for OPTIC, represented by square markers and a solid line, and another for POPF, represented by circular markers and a dashed line. OPTIC generally shows a higher computation time than POPF at smaller instance scales but grows more gradually. POPF starts with a lower computation time at S1 but increases more steeply as the instance scale grows. Key data points include OPTIC's computation time starting at around 21.83 seconds at S1 and reaching up to approximately 1000 seconds at VL3, while POPF starts at around 0.28 seconds at S1 and also reaches up to approximately 1000 seconds at VL3. All values are approximated.Average computation time per instance for disrupted-operation instances (VL3 averaged over its nine solved instances). POPF is faster on small instances (0.28 s at S1 against 21.83 s for OPTIC) but its runtime grows more steeply with scale