SimWell built a dynamic simulation model for a gas distributor in Western Canada representing their compressed natural gas distribution system in North Dakota. The objective of the project was to size the vehicle fleet to minimize costs while ensuring a 100% level of service.
Complexity of the project
Sources of supply and customers are continually on the move because gas is produced and consumed by petroleum well drilling activities. In addition, compressed natural gas is stored at the customer’s premises in the distributor’s trailers. Finally, the variations of the weather, the maintenance of the equipment and road blockages influence the achievement of the desired level of service.
The complexity of the distribution network exceeds the capacity of mathematical tools. Therefore, a discreet simulation model was built using Arena Simulation Software from Rockwell. Since suppliers and customers are constantly on the move, it is impossible to predict the precise location where they will be located in the future. Therefore, SimWell has to create a grid of Horizontal coordinates (longitude, latitude) on a road map of the area served. A travel calculation software was used to calculate travel times between sites and a Markov chain was used to represent the possible impact of the weather on travel times. Finally, SimWell created an algorithm to manage the dispatching of tractors and trailers.
The model developed by SimWell allowed its customer to size its fleet of tractors and its fleet of trailers, in addition to testing the operating processes to choose the most efficient one. The customer has therefore minimized the risk of falling short on the required level of service demanded while keeping its costs at the lowest.
Here is a video showing the challenge braved by Ferus in North Dakota: