In this paper we study a tour-scheduling problem for an attended home delivery problem with uncertain order requests. The problem is modelled as a two-stage stochastic programming problem and solved using the multi-cut L-shaped method. Weekly working patterns are generated by means of a pricing heuristic. Numerical results on randomly generated instances show that including weekly working rules increases the total cost by only a small amount (up to 2.54%) when compared to an approach that only considers daily working rules for the generation of driver schedules.