Purpose -The purpose of this paper is to develop a method to evaluate the quality of service delivery process designs, based on how closely they meet process requirements of key stakeholders while taking variability into account. Design/methodology/approach -A Monte Carlo computer simulation of the flowchart of the service delivery process is used to capture the effects of multiple types of variabilities and parametric uncertainties on process variables of interest, and the Taguchi quality loss framework is applied to estimate overall process quality. As an example, a proposed modification to a patient-treatment process in a hospital emergency department is evaluated. Findings -This paper demonstrated a method that service managers can use to evaluate the quality of service delivery process designs. Practical implications -This method can evaluate modifications to various aspects of a service delivery process, and can assist managers to fail-safe the entire process. Furthermore, it characterises process quality using a range of probable values instead of a single value, such as the process mean, and thus provides a more realistic representation of quality. Originality/value -This paper contributes to service engineering in two ways: by adapting the Taguchi quality loss approach, commonly used to assess manufacturing process quality, to a service delivery process; and by incorporating uncertainty, due to imprecise knowledge of process parameters, into the quality assessment.