A review of key planning and scheduling in the rail industry in Europe and UK (original) (raw)
Related papers
Scheduling and Rescheduling of Railway Operations: A Review and Expository Analysis
Technology Operation Management, 2011
Railway operations are planned in four levels of hierarchy: (i) strategic, (ii) tactical, (iii) operational control and (iv) real-time control. Scheduling is an initial time allocation of resources to meet demands in completing a task and rescheduling is a later modification of such resource allocations. An extensive set of some major railway scheduling and rescheduling operations as published in literature is analyzed in this paper. We observe that many authors consider scheduling as a strategic or tactical operation and rescheduling as an operational or real-time control. Several unexplored issues that are strategic or tactical per se, but hold considerable rationale for operational and real-time control (and vice versa) of railway operations are brought out. It is interpreted that (i) in practice, both scheduling and rescheduling are inadvertently performed at all hierarchical levels and (ii) operational feasibility at a hierarchical level is dependent on effectiveness of modeling at other hierarchical levels. Finally we present an array of potential research issues that are significant in practical railway operations. Our discussions are based on an understanding of objectives, modeling details, approaches and issues arising from the diversity in railway operations and to a less detail on solution methods and techniques.
A decision support system (DSS) for the railway scheduling problem
Artificial Intelligence …, 2004
The recent deregulation occurred in the public railway sector in many parts of the world has increased the awareness of this sector of the need for quality service that must be oered to its customers. In this paper, we present a software system for solving and plotting the Single-Track Railway Scheduling Problem eciently and quickly. The problem is formulated as a Constraint Satisfaction Problem (CSP), which must be optimized. The solving process uses dierent stages to translate the problem into mathematical models, which are solved to optimality by means of mixed integer programming tools. The Decision Support System (DSS) we present allows the user to interactively specify the parameters of the problem, guarantees that constraints are satised and plots the optimized timetable obtained.
CREWS: A Train Scheduling Tool
WIT Transactions on the Built Environment, 1998
We present a tool, CREWS, for generating applications in personnel scheduling in transportation companies. Systems based on CREWS have been developed for the Portuguese Railways, the Dutch Railways, the Norwegian Railways, and the West Anglia Great Northern Railway (one of the operators of the British Rail). The initial version of CREWS supported long-term scheduling of drivers and guards, but current developments are extending it to support short-term planning as well. CREWS heavily relies on the use of Artificial Intelligence techniques and has been built in the perspective of a "white box" system, in the sense that the planner can perceive what is going on, can interact with the system by proposing alternatives or querying decisions, and can adapt the behaviour of the system to changing circumstances. Scheduling can be done in automatic, semi-automatic or manual mode. CREWS has mechanisms for dealing with the constant changes that occur in input data, can identify the c...
Applications of expert systems in railroad maintenance: Scheduling rail relays
Transportation Research Part A: General, 1990
Railroads replace rail when it exhibits too many defects, wears beyond safe limits. or develops extensive plastic deformation. The complex process of developing a plan for rail relay or replacement is known as rail scheduling. A knowledge based expert system has been developed to guide the rail scheduling process. A prototype was developed in conjunction with Burlington Northern Railroad (BN) in order to provide more formal, more consistent, and more rational criteria for rail replacement. A productive system was then developed for use in preparing BN's 1990 rail relay program. The system demonstrates the role of expert systems in supporting decision making in railroad maintenance and identified several areas for other applications of expert systems.
Algorithmic support for human rail shunting planners
Proceedings of the 18th …, 2005
Traditionally, algorithms that generate schedules are either based on domain analyses or on task analyses. Both approaches have disadvantages. Algorithms based on domain analyses are often not used by human planners and schedulers. Algorithms based on task analyses, however, do not use the potential improvements in quality that domain based algorithms can attain. In this paper, the advantages of both approaches are combined by applying a mixed approach. Combinatorial optimization algorithms have been developed and implemented for two subtasks that were identified in a task analysis of shunting planners at The Netherlands Railways. Because of the approach taken the algorithms are usable in practice, which can improve both the quality and the speed of generating plans.
An Industrial Engineering Approach to Improve Railway Planning: A case study
Railway planning is a continuous and very complex planning process in which many planners are involved. The process of plan creation can be characterized by long throughput times, strict division of tasks, a high task complexity, and a lack of standardization. The resulting plans need to be robust and satisfy all kinds of safety regulations. Literature has given attention to shunting task planning, but often the proposed solution approaches have not been implemented in planning systems. We investigate reasons for the lack of applicability of such solution approaches. Next, we examine the process of creating plans for shunting operations at stations during the night in a case study. Several planners are involved in this process. Both the efficiency and effectiveness of this process had to be improved. We apply an industrial engineering approach to improve this process. This approach consists of a process analysis, a task analysis, an investigation of improvement possibilities, and finally the development of computerized tools for improving parts of planning tasks. The result is an improved system for supporting the planning process. We will discuss characteristics of this support system and the implemented algorithms.
Railway Station Planning Using ArtificialIntelligence Techniques
WIT Transactions on the Built Environment, 1970
The paper deals with a new method to outline by means of using artificial intelligence techniques the optimal scheme of a railway station plan which can satisfy a known O/D traffic matrix. This method assumes as inputs of the choice model the matrix of line changes, that is the number of trains which must be scheduled from one line to another in a station and over a known interval of observation. It respects safety constraints and its aim is to obtain the maximum capacity for circulating train. Capacity must be calculated taking into account all the elements of the scheme of a station. A series of applications shows to what extent this method is valid and extendible and suitable to any other application.
IEEE Transactions on Systems, Man, and Cybernetics, 2001
Predictive and reactive train scheduling are tactical and operational decision making, respectively, under constraints (e.g., resource capacity, managerial objectives) and under uncertainty (e.g., imprecise data and information, unforeseen events). Predictive scheduling produces timetables taking into account the market demand and resources utilization levels. Reactive scheduling challenges disruptions to timetables and schedules trains and operations with imprecise plans. Expert knowledge is indispensable for finding practical solutions for both predictive and reactive scheduling. Consequently, knowledge management (KM) strategies, processes and technologies can improve the decision-making process and outcomes. This paper focuses on the following issues. Five dimensions are introduced to distinguish predictive and reactive train-scheduling activities. The combined use of data and knowledge and the differences in uncertainty levels are used to position comparatively the two scheduling approaches. The intensity of reliance on explicit and tacit knowledge is highlighted via the elaboration and classification of knowledge used in either one or both scheduling environments. The significance of train-scheduling tacit knowledge elicitation is described by, first, presenting a real case analysis which resulted in the elicitation of rich and valuable tacit knowledge (timetabling heuristics) from explicit knowledge (timetable) and, second, generalizing lessons learned from this process. The contributions of the tacit knowledge elicitation process to the enhancement of the train-scheduling system which leads to better resource utilization and customer satisfaction are itemized.
A decision support system for railway timetabling (MOM): the spanish case
Computers in Railways …, 2006
Train timetabling is a difficult and time-consuming task, given the complexity and number of constraints to be taken into account. A feasible train timetable should specify the departure and arrival time of each train to each location of its journey so that the line capacity and other operational constraints are met. Traditionally, train timetables have been generated manually by drawing trains on the time-distance diagram, where train schedules are manually adjusted so that all constraints are fulfilled. This process can take a long time and it usually stops once a feasible timetable is found. The difficulty of the process increases in complex and high-loaded networks and the resulting timetabling may be far from optimal. In this work, we present MOM: a friendly, flexible, computer-based Decision Support System (DSS) whose main goal is to support the railway planner to efficiently obtain optimized train timetables. The system generates feasible and high quality timetables by considering the set of constraints of the problem and optimization criteria related to the use of railway infrastructures and operator requirements. The railway planner can obtain several solutions automatically, by modifying certain parameters (departure interval, frequencies, among others). Hence, the user can analyze them and take decisions about conditions related to railway traffic and infrastructure. Furthermore, an important feature of the DSS is the consideration of hard and soft constraints which makes the search process more flexible and reaches better solutions. The developed DSS offers assistance in train scheduling, conclusions about the maximum capacity of the network, identifying bottlenecks, determining the consequences of changes, providing support in the resolution of incidents, and alternative planning in on-line scheduling for real traffic control. This DSS software is being successfully used by the Spanish Manager of Railway Infrastructure (ADIF).