This article was published in the October issue of International Cement Review in 2015.
The use of algorithms in transport planning software can cut cement distribution costs and increase customer service levels at the same time. But not all algorithms are the same and the level of quality can mean the difference between good or optimized planning performance. A look under the bonnet explains how they work.
■ by Thomas Bergmans and Ulrich Dorndorf, INFORM GmbH, Germany
The power and quality of transport planning
Planning the supply of cement from a plant to either storage depots and customer sites can be challenging. Planners face a high variability of demand, heavy onsite traffic during peak hours, limited availability of trucks, driving time regulations and many other restrictions. At the same time, they must try to find the best balance between customer satisfaction and transport costs. Today, deliveries from cement works are often planned independently on a plant-by-plant basis to serve the demand within a facility’s catchment area.
By using real-time optimisation software for network-wide transport planning, resources can used in a more efficient way. In a smart network, vehicles can be shifted where required, preventing idle times in some locations and work overloads in others. This approach has recently been successfully implemented at one of Germany’s largest cement suppliers.The two basic elements for such an optimisation approach are an existing corporate-wide enterprise resource planning (ERP) system and an intelligent optimisation software for planning and controlling transport.
The optimisation approach to planning
Transport optimisation software can be used as an add-on to an existing ERP system, following a best-of-breed strategy for selecting the solution that is most suited for a particular task. Optimisation benefits from the enormous advances in computer hardware but much more from even greater strides that have been made in algorithms and software over recent decades.
Optimisation approaches are commonly based on formal decision models. The goals and constraints are explicitly stated, and the user can influence them through configuration. They implicitly define the possible courses of action. What remains is the task to select the best course from the huge set of possibilities.
Continue reading - request the full article using the form below.