Multipurpose optimal designs for hypothesis testing in normal response trials