On the Design and Implementation of a Process Modelling Language for Uncertainty

A modelling framework focusing on the treatment of uncertainty is proposed. The framework is based on Julia, given that it is an open-source, modern, computationally efficient language (scripted but with just-in-time compilation) providing support for the design of domain specific languages via abstract types, multiple dispatch, operator overloading and full Unicode support. The framework provides a concise and natural syntax, allowing for a traditional mathematical notation. For instance, users can write t = 700 +/-(3 delta) (uncertainty range), X similar to N (mu, sigma(2)) (probability distribution) or F-0 = V-3 +[21.3,21.7] (interval arithmetic) without having to compromise between speed and code readability. A case study on the scale-up of an Aerosol-Assisted Chemical Vapour Deposition (AACVD) process is presented to exemplify the use of the framework.