Real-Time Optimisation of Empty Rail Freight Car Distribution
by Martin Joborn
The rail freight carrier Green Cargo has a new system for rail freight car distribution planning. It is an automatic, real-time planning system, based on a kernel of optimisation models and methods. The new system has resulted in both significant savings and quality improvements.
In railway systems, it often occurs that there is an imbalance in the freight flows, meaning empty freight cars must be moved in order to enable new transports. An efficient empty freight car distribution process is vital for good utilisation of the freight cars. A bad distribution process can result in delayed transports, good-will losses and an unnecessarily large car fleet with the associated capital and maintenance costs.
|Full control over freight car planning at Green Cargo. Photo: Peter Lydén / Green Cargo.
Green Cargo is the largest railway cargo carrier in Sweden. In collaboration with Carmen Consulting and Linköping University, Green Cargo has developed an advanced new system for planning their empty freight car distribution. The aim of the system is to improve the quality of the empty freight car distribution process. The system handles about 7000 freight cars, divided into 25 types. The railway network has about 1500 pick-up and delivery points for freight cars, and the system produces detailed transportation plans between these points.
The kernel of the new system is an optimisation model that calculates distribution plans in real time. The model is constantly updated and re-optimised as new information becomes available, eg updated freight car status, transport orders or train capacities. The response time of the system for re-optimising the plans is less than one second. The plans always reflect the optimal distribution for the current situation. The objective of the distribution planning is to maximise the number of filled transport orders while minimising transportation costs, as well as to generate distribution plans with a preferred structure and to ensure that cars arrive punctually to customers.
The high performance of the system makes it possible to respond immediately to customers' transportation demands. While on the phone placing the transport order, the customer will also be informed of whether freight cars are available for the transport. The distribution system generates an online optimisation to calculate and guarantee the delivery of the empty freight car to the transport. However, regarding the method of satisfying a customer's freight car demand, the system maintains flexibility for as long as possible, while simultaneously ensuring that the order will be filled.
The system is fully automatic and needs no manual interaction to execute the calculated plans. At the correct time, stations, terminals and operating crews receive instructions relevant to their cars and transport orders - bookings, shunting orders etc - from the system.
The introduction of the new system has made empty freight car distribution a trustworthy and highly precise process, which has lead to loyalty from both personnel and customers. The reliability of the process decreases the need for backup solutions such as local freight car inventories. Furthermore, the freight car controllers have a good overview of the freight car supply-and-demand situation over the system's planning period of one week.
The improved utilisation of freight cars has enabled Green Cargo to reduce its car fleet and the associated capital and maintenance costs. Green Cargo estimates its yearly savings at €7.6 million, excluding the benefits from improved customer relations.
The concept used at Green Cargo is applicable to other problems as well. Interesting potential applications include loaded freight car planning, taxi planning, truck allocation, and container distribution.
Martin Joborn, Carmen Consulting, Sweden
Tel: +46 8 555 79 438