The issue all of us face in using the BEICF data, is lack of structure and mechanism to quantify if controls fail. In most cases the implication of control failure and linkage to data loss is absent. So while mutually exclusive data exists, the usge to adjust AMA becomes a judement call rather than a statistic method of impact calculation One Risk Professionals suggests that "One of the possible mechanisms of using BEICF to adjust AMA operational risk capital is to quantify operational risk framework components (loss data, RCSA, KRI, action plan, etc) to arrive at business unit risk profiles using scorecards. You could adjust model capital numbers based on relative movement of risk profiles over a period of time." There is indeed a lack of structure in BEICF data and the manner in which it could be employed in computing operational risk capital using statistical models. One way is to possibly use the performance of selected operational risk framework components in influencing the scenario data parameters that are used as an input in the capital computation process. The operational risk capital is computed at the group level and allocated to business lines using the variance-covariance method. Post this the capital that has been allocated to business lines can be adjusted using scorecards that capture BEICF performance. Thus the capital is adjusted to reflect current risk management practices rather than being based on historical data only and it also serves as a mechanism to incentivize business line managers to improve their control standards and risk management practices