A plan of military air operations contains a number of sequential and/or pa
rallel tasks. Given information of the battlefield situation (including the
status estimate of tasks) during the execution of a plan, it is necessary
to assess the situation impact on the plan, especially whether the entire p
lan would be delayed. Plan performance assessment fuses the status estimate
of tasks into the status estimate of the entire plan. The performance asse
ssment of the plan assists the commander of air operations to decide the ne
ed for plan modification to assure the goal of air operations. The performa
nce assessment of the plan must deal with uncertainty in the status estimat
e of tasks in the plan. This paper presents a comparative study of two info
rmation fusion techniques that infer the status estimate of a plan from the
status estimate of tasks under uncertainty: the decomposition technique wi
th low computational cost and the composition technique with high computati
onal cost. The testing results of these techniques indicate the similar est
imation accuracy of two techniques. Due to its low computational cost, the
decomposition technique is recommended. Guidelines for applying the decompo
sition technique to a large-scale, complex plan of the tasks are also provi
ded.