Reasonably foreseeable direct and indirect misuse can contribute to unreasonable risk. Direct misuse is analysed within the triggering condition analysis; insufficiencies that could defeat a measure against indirect misuse are also in scope. Causes include users not understanding the system or being misled by similar systems with different rules, wrong expectations from poor information, loss of concentration, overreliance and wrong assumptions from the design. Methods include known misuse from field experience, studies with test subjects, use case and interaction analysis, HMI analysis, known patterns of misuse and automation complacency, human capability to regain awareness, and applicable standards and guidelines. Driving while incapable of the driving task is abuse and out of scope. The effectiveness of misuse measures can be shown during V&V.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.