Proefschrift
also known as a switching manifold. The stopping criteria for this iterative algorithm is determined by finding the relative error of the control variable and requiring that it be less than a tolerance δ = 10 − 4 . In this way, the algorithm terminates when Λ ◦ ( n − 1) − Λ ◦ ( n ) Λ ◦ ( n ) < δ (B.8) where · is the 1 -norm, Λ ◦ ( n ) = t f j =1 | Λ( j ) ◦ ( n ) | . The initial guess of the optimal treatment is given by Λ ◦ (1) = 0 . 5. The vector Λ ◦ ( n ) is then used as the optimized treatment protocol. B.3 All Optimized Treatments The Forwards Backwards Sweep algorithm is used for each of the six stability points provided in Table B.1. The optimal abiraterone dosing schedule to arrive at the stable point from each of the 100 starting points is shown in Figure B.5 - Figure B.10. Figure B.5: Matrix 2 Forward Backwards Sweep results for optimal dosing schedule to arrive at two-species stability point (2082 . 76 , 5206 . 90 , 0 . 00). Figure B.6: Matrix 2 Forward Backwards Sweep results for optimal dosing schedule to arrive at three-species stability point (863 . 45 , 4436 . 73 , 694 . 82). 143
RkJQdWJsaXNoZXIy MjY0ODMw