Examines the Reliability Improvement Warranty (RIW). Consumer and commercial warranty experience does not alone justify optimistic expectations for RIWs; the effect of RIWs in completed Department of Defense (DOD) programs is inconclusive; and as a result of inadequate research design the expectation of drawing meaningful conclusions from the ongoing RIW experiment may be over-optimistic. Examination of completed RIW programs, however, suggests the importance of (1) modification after operational use or testing, (2) schedule flexibility, (3) contractor involvement in initial overhaul and repair, and (4) avoidance of RIWs in programs subject to extreme quantity or utilization uncertainty. The design of the DOD's ongoing RIW experiment can be improved by (1) reducing the variation in contractual terms, (2) developing better controlled conditions, and (3) establishing defined limits for the experiment. In addition, the DOD must recognize the multiple objectives of the RIW and establish priority among them to facilitate evaluation.