Fuzzy calculus applied to real time scheduling

Real time systems are more and more complex and, increasingly, involve large applications where real time constraints can be only roughly defined by use of vague linguistic terms. Generally, real time scheduling algorithms provide optimal solution when the application context is well known and its behavior completely mastered. But in practice, both the real time constraints and the tasks characteristics are imprecisely known. This results in suboptimal behaviors generating failures which should be avoided. This paper proposes to apply the fuzzy calculus to real time tasks scheduling in order to allow both more realistic knowledge representations and definitions of more flexible real time scheduling algorithms