Pervasive computing applications involve information flow across multiple organizations. Thus, any security breach in an application can have far-reaching consequences. However, effective security mechanisms can be quite different from those typically deployed in conventional applications since these mechanisms are constrained by various factors in a pervasive environment. In this paper, we propose a methodology to perform a cost-benefit analysis under such circumstances. Our approach is based on the formulation of a set of constrained multi-objective optimization problems to minimize the residual damage and the cost of security provisioning. We propose the use of workflow profiles to capture the contexts in which a communication channel is used in a pervasive environment. This is used to minimize the cost that the underlying business entity will have to incur in order to keep the workflow secure and running.