The Canadian Pacific Railway Transforms Operations by Using Models to Develop Its Operating Plans

Abstract

North American railways have traditionally practiced tonnage-based dispatching, running trains only when they have enough freight. As a result, their customer service and their use of crews, fixed assets, locomotives, and railcars are poor. Canadian Pacific Railway is using new decision-support tools developed in-house and by MultiModal Applied Systems to create a scheduled railway. These tools use operations research approaches, such as an optimal block-sequencing algorithm, a heuristic algorithm for block design, (very fast) simulation, and time-space network algorithms for planning locomotive use and distributing empty cars. This implementation has saved $300 million Canadian (US$170 million) from mid-1999 through autumn 2000. We estimate it has saved at least an additional $210 million Canadian during 2001 and 2002 in fuel and labor costs alone. Labor productivity, locomotive productivity, fuel consumption, and railcar velocity have improved by 40, 35, 17, and 41 percent, respectively. Furthermore, Canadian Pacific Railway now provides its customers with reliable delivery times and has received many customer and shipping association awards for its improvement in service.

Study of the State of Logistics in Kazakhstan: Prospects for Development and Deployment of Transport and Logistics Centres

“…Network solutions of the Canadian Pacific include seaports associated with roads and railways (east and west coast of North America) and 145 intermodal freight terminals, transhipment points, and freight yards [20]. The network of transport and logistics centres of the Canadian Pacific is linked to the centralized control system and single technological process and provides connectivity of territories of the US and Canada, as well as access to key maritime trade routes of Southeast Asia and Europe direction.…”

“…In terms of the most important global trends for Kazakhstan, the following has been highlighted: 1) development of a multi-level and effectively managed network of internal and external transport and logistics centres as a cargo management system on a vast geographical area (Germany) [17]; 2) development of the concept of 'dry ports' near major producers and consumers (USA, Mexico) [12; 18; 25]; 3) creation of transnational multimodal integrators on the basis of the largest companies that perform the task of accelerated development of the logistics part on the vast geographical area (Germany, Canada) [20]; 4) introduction of the principles of supply chain management and contract logistics; 5) improving the efficiency of customs procedures (Singapore). For example, the establishment of an efficient customs has become one of the key factors in the development of Singapore as a global trading hub; 6) raising the level of containerization; 7) introduction of innovative and information technologies; 8) creation of attractive conditions of service of transport companies; 9) tendency to improve the quality of logistics services along with growing competition; 10) development of networks of airports and high-speed railway lines for passengers.…”

Train design and routing optimization for evaluating criticality of freight railroad infrastructures

“…The congestion cost at links and classification cost at yards are considered in CN t ij if the block is carried by train t from yard i to yard j. These calculations are very different from the shortest-path algorithms used by MultiModal (Ireland et al, 2004) on the original physical network.…”

“…Cacchiani et al (2014) recently provided an overview of real-time models and algorithms for recovery from disruptions or disturbances. In practice, North American railroads widely use the MultiModal system to create the sequence of blocks and train routes with both optimization and simulation techniques (Ireland et al, 2004). MultiModal creates block sequences by minimizing the number of switches and total car-miles under various constraints through shortest-path algorithms.…”

The locomotive assignment problem: a survey on optimization models

“…The schedule-based strategy implies that the schedule should be adapted depending on the forecast of the traffic and requires advanced computers and operations research tools to conduct deep analyses of different alternatives in short times. As reported in Ireland et al (2004), the schedule-based strategies recently have gained favor in the United States and Canada, where several railroad companies have adopted this more disciplined approach to obtain cost-effective and customer-effective operating plans. The increase in customer demand for freight rail transport and the recent availability of advanced computers and OR software have pushed several North American railways to change the paradigm of their operations, passing to a schedule-based strategy.…”