Currently many American industrial firms are considering the adoption of Just-In-Time (JIT) as an inventory control/material flow system. A JIT system can have several configurations. Examples exist of JIT being applied system-wide between all the echelons which make up a channel. There are also examples of firms adopting JIT only on the materials management side, or alternatively, only on the physical distribution side of a particular focal organization (echelon) within a channel. The reality of uncertainty in the channel's operation and environment suggests that alternative inventory systems (such as JIT) must be evaluated under a range of internal operational uncertainty and external demand uncertainty conditions. This study offers a rational methodology to anticipate the performance impact of alternative system designs under realistic operating conditions. The G.P.M. simulation model was used to represent the operation of a channel under sixteen treatment conditions. The research design was full factorial with two levels each of four factors (physical distribution JIT, materials management JIT, materials management operational uncertainty, and demand uncertainty). The response variables which were used as indicators of channel performance included: profitability, order cycle time, standard deviation of order cycle time, and percent customer orders filled. Eleven research hypotheses explored the relationship between JIT configuration and uncertainty, both in terms of profitability and physical distribution service level maintained. The major conclusions of the research are: 1. JIT is not the unidimensional system often depicted in descriptive studies. JIT effects tend to be complex, interactive, and level dependent. It is particularly difficult to predict the effect of JIT on one echelon, or on subsystems within echelons. 2. Rather than the inherent positive effects often attributed to JIT, results indicate negative effects for both profit and service under a range of uncertainty conditions. 3. Results support the common criticism of JIT that its performance is sensitive to uncertainty, particularly demand uncertainty. The performance of non-JIT systems were also shown to have similar sensitivity to uncertainty. 4. Most synergistic interactions between factors were not significant, but the statistical procedure for means comparison was acknowledged to be conservative. 5. Results also indicated that JIT systems may make the job of maintaining high customer service levels more difficult. / Ph. D.