In this paper we propose a multi-scale approach together with integration strategies for a continuous and dynamic planning and scheduling decision problem from the process industry. The decisions have to be made before all data are available in contrast to most sophisticated planning and scheduling approaches for the process industry that consider a fixed time horizon and assume that all data is given at the time of application. The approach is based on a hierarchically structured moving horizon algorithm. On each level we propose optimsation models to provide support for the relevant decisions. The levels are integrated with versatile integration strategies that transfer and implement the decisions at the adjacent levels. The algorithm based on the integration strategies restricts the solution space to eliminate infeasible solutions and uses hard constraints, bounds, shaping methods and penalty functions as guidelines, for obtaining near-optimal solutions. Feasible solutions can still be obtained when the guidelines are violated although they become less optimal. Solution procedures have been developed and the integrated multi-scale approach has been validated and tested with data from the real world problem.