A multinational company in the area of agrochemicals experienced a loss of profits for five consecutive years for a family of products belonging to its portfolio, due to increasing raw material prices, creating significant pressure on its supply chain. This work concerns an industrial case study with two main objectives, addressed in two different stages. In the first stage we redesign the worldwide supply chain network and optimise the production and distribution planning over a time horizon of one year (corresponding to the seasonal demand cycle for final products), while providing a decision support tool for long term investments and strategies (capacity expansions, production outsourcing, new manufacturing and storage locations, etc). The robustness of the first stage decisions are then tested in a second stage detailed planning model. The company has a major manufacturing site in a given country where all active ingredients (AI) are produced to supply a worldwide network of formulation sites, as well as some intermediate and final products to supply local and international markets. The worldwide network of formulation sites consumes AI and produce final products to supply the global (international) customers' network. The company uses a network of public warehouses and terminals to distribute the final products to customers in local and international markets. Customer end products may come in several forms: formulated or technical, bulk or packed products, etc, which translates into an extensive portfolio to manage. The structure of the network is not rigid; the number of production stages depends on the product being manufactured and the final products can be sent to final customers from any point of the supply chain. In this stage, the objective is the maximisation of the net profit value (NPV) of the company, and an aggregate description of resources availability and production tasks is adopted, i.e. no changeover operations are considered. The output decisions from the first stage, mainly the supply chain configuration and product/site or customer/warehouse allocation decisions, are used as input parameters in the second step. In this stage, a short term planning model is developed, where a set of individual orders placed by customers has to be fulfilled in time. The outputs are detailed production and distribution schedules as well as an assessment of the network responsiveness, i.e. service level in final customers. Usually, failure to meet demand in time, when soft orders are placed, is caused by allocation of too many products/ customers to the same resource in the first stage, creating local bottlenecks, even if manufacturing/ distribution capacity is available in other locations. Some of these bottlenecks only become problematic when performing the detailed production/ distribution scheduling. This happens due to the utilisation of a single optimisation criterion (NPV maximisation) and the aggregate description of manufacturing activities in the first stage. An algorithm was developed to use the information gathered in the second step to improve the supply chain design by enforcing a more distributed allocation of products/ customers to the available resources and reduce the number of changeover activities taking place at each time period. The responsiveness testing is performed under a wide range of scenarios in order to guarantee the robustness of the derived network in the first stage.