Room: Karl Dean Ballroom A1
Purpose: New techniques of acceptance testing and commissioning (ATC) has been introduced for a new class of ring-gantry linac. The procedures relies heavily on MV portal images and automated analysis. The design of the machine and the display console differs from the conventional c-arm (Varian 2300EX/Trilogy) systems. We performed a risk assessment for implementing the new ATC procedures for the pre-commissioned system.
Methods: Instead of relying on the beam data provided for the pre-commissioned machine, additional measurements were performed to supplement the spot-checks recommended by the vendor. Five physicist, who were involved in the ATC of 3 machines (1 academic and 2 veterinary clinics), participated in the failure modes and effects analysis (FMEA). Potential failure modes, causes of failure, and their impact were identified. A process map was generated and the probability of occurrence, severity, and detectability for each failure mode (FM) was scored to generate the risk priority number (RPN) following AAPM-TG100 guidelines.
Results: A total of 9 sub-processes (with 15 potential FMs) and 19 sub-processes (with 45 potential FMs) were identified for the acceptance and commissioning process respectively. The top 3 failure modes for acceptance includes: error in absolute dose calibration (RPN = 118), machine temperature and water flow rate inconsistencies (RPN =112), and insufficient door clearance to fit the assembled machine (RPN=101). The top 3 FMs for commissioning includes: error in determining the non-isocentric SSD (e.g. SSD of 90cm) from lateral MV images (RPN =128), error in determining the SSD due to the dependence on the window setting of the image intensity (RPN =122), and the potential error for setting up unique calibration options recommended by the vendor.
Conclusion: Potential FMs and hazardous steps in the ATC process have been identified. This analysis will be invaluable in mitigating errors and ensuring that the ATC is completed correctly.