This invention relates generally to railyards, and more particularly to locomotive parking and servicing management within a railyard.
Most railyards must store incoming locomotives between assignments to trains, and many railyards also carry out service operations on locomotives. Both the parking and/or servicing of locomotives can affect the time at which they will be ready for service on an outbound train, so parking and service decisions can materially affect the overall performance of a railyard. In general, it is recognized that railyard management would benefit from the use management tools based on optimization principles. Such tools use a current railyard status and list of future tasks to be accomplished to determine an optimum order in which to accomplish these tasks such that railyard management objectives and rules are fulfilled.
As used herein, the term “locomotive consist” or “consist” means one or more locomotives physically connected together, with one locomotive designated as a lead locomotive and other locomotives designated as remote locomotives. The term “train consist” means one or more locomotives and one or more railcars physically connected together.
Railyards must store locomotives temporarily, when inbound or terminating trains are disassembled. The locomotives are parked in the yard, and placed back into service later as needed. Many yards include a locomotive service shop, and inbound locomotives therefore fall into one of four classifications: assigned to a later outbound train, needing no service, unassigned, and needing no service, assigned to a later outbound train, and needing service, and unassigned, and needing service. Depending on the locomotive's status and the schedule of inbound and outbound trains, a given locomotive may need to remain in the yard for a short while, or for a long time. The parking arrangement for locomotives should, if possible, accommodate the easy retrieval of locomotives at the time they must be moved, but limited parking facilities generally complicate the situation.
A typical parking arrangement for a railyard, comprises a collection of parallel tracks and a locomotive shop, located side-by-side. There is usually a direction of flow through the railyard with locomotives normally arriving at the parking complex, and later being pulled for service from the parking complex. However, an arriving locomotive will frequently be parked behind other locomotives, and if it is needed before one of those which precede it in the queue, then additional locomotives must be temporarily displaced in order to free the needed one. This represents an inefficiency, both in terms of time delay and labor hours needed to perform the extra activities.
Another inefficiency arises if locomotives slated for service are parked in a poor order. For example, a locomotive requiring 30 minutes of service, and slated for outbound use three hours later may be parked behind a locomotive requiring four hours of service. In order to meet schedule, the obstructing locomotive must be moved, again resulting in delay and cost in hostler hours.
There exists a need for a locomotive parking management scheme to ameliorate the inefficiencies which arise in any given parking/service configuration. As locomotives arrive, there will be several options available for parking them, either for use or for service. A desirable parking management scheme is one which is capable of weighing the cost of various parking options against the future locomotive requirements of the yard.
In one embodiment, a system for managing locomotives in a railyard including a storage or parking yard and a service yard, determines an optimal configuration of locomotive within a railyard, based on possible future states of the parking yard and the service yard. The system includes a computer and utilizes an algorithm that enumerates possible present locomotive placement options, enumerates possible future railyard states arising from each possible present locomotive placement option, examines each possible future railyard state, and determines a present option based on the examination of the possible future railyard states.
More specifically, the system establishes an initial state of the railyard by evaluating a geometry of the parking yard and the service yard, and evaluating a present configuration of locomotives in the parking and service yards. The system then enumerates possible future railyard states based on evaluation of the initial railyard state and a yard schedule, which includes an inbound locomotive schedule and an outbound locomotive schedule. Additionally, locomotive service requirements and non-standard movements are considered when enumerating possible future railyard states. Next the system examines each possible future railyard state wherein a cost and a time-based efficiency of each possible future state is calculated. The cost and efficiency calculation considers the effect of railyard operations such as, the cost and time delay caused by locomotive service requirements, and the cost and delay of non-standard locomotive movements.
Finally, the system determines a present locomotive placement option by applying specific railyard locomotive management objectives and rules, and selecting the present placement option that will provide future states that most closely fulfill the management objectives and rules. The management objectives include such things as, assembling an outbound train as scheduled, delivering the outbound train as schedule, reducing labor involved in assembling and delivering the outbound train, and reducing delays in locomotive servicing.
The management rules include parking yard management rules such as executing locomotive pull-forwards when there is a reduced number of locomotives on an affected track, maintaining an order of locomotives on each parking track such that locomotives for later outbound locomotive consists are parked behind locomotives for earlier outbound locomotive consists, and parking a lead locomotive for an outbound locomotive consist on a parking track such that the lead locomotive is in front of other locomotives parked on the same track that are allocated for the same outbound locomotive consist. Additionally, the management rules include service yard management rules such as positioning a locomotive in a queue for service on a lead-in track to a service bay that provides the appropriate service, positioning locomotives in queue on a lead-in track in an order that allows servicing of each locomotive to be completed before each locomotive is scheduled for assembly in an outbound locomotive consist, and scheduling short service activities before long service activities when scheduling conflicts are not at issue.
Thus, the system enumerates possible present locomotive placement options, examines possible future railyard states that result from each option, and processes incoming locomotives based on the placement option having future states that fulfill the railyard management objectives and rules.
In one embodiment, locomotive managing system 10 manages locomotives in a railyard based on possible future states of the yard. To begin, system 10 (shown in
Thus,
As inbound train consists arrive at the railyard locomotive parking decisions must be made. In one embodiment, the locomotive parking process utilizes the following guidelines,
Given the initial state of system 10, the inbound schedule, the outbound schedule, and the parking options, the locomotive manager is confronted with providing parking and facilitating service, as needed, for all locomotives present in the yard, and doing so in a manner which meets the following locomotive management constraints,
Referring to
Typically, locomotive parking does not follow an ideal FIFO (first-in, first-out) flow through parking yard 52 and service area 60, and the parking arrangement, at the expense of extra man-hours of labor, is not ideal. For example, at some extra expense in labor and time, an inbound locomotive may need to be pulled around parking tracks in parking yard 52 and parked at the front of a parking track. Such a move might well justify the extra cost if in fact that locomotive is needed before some of the other locomotives already on the same parking track. Furthermore, to avoid extra labor costs for arranging the order of a power consist, a lead locomotive is best placed in front of the other locomotives for the same outgoing train.
However, there are unavoidable minimum labor requirements for moving locomotives to parking yard 52, and placing them in the input end of the parking or service tracks. Thus, when deviations occur from the FIFO order the parking management process trades off costs of alternate parking arrangements. Such out-of-the-ordinary moves as referred to as non-standard moves (NSM's). Each of NSM has a cost in man-hours of labor, based on the actual yard geometry. In one embodiment, the following actions are regarded as NSM's, and subject to extra costs,
In addition to the initial state, the inbound schedule, the outbound schedule, and the parking options, the parking management algorithm utilizes the cost in man-hours and dollars associated with NSM's, delays associated with NSM, cost in man-hours and dollars associated with the delays, a list of service types provided by the diesel shop, and a cost in dollars associated with an outbound train consist not departing on time.
After the initial state, yard schedule, costs of NSM's, and the other information utilized by the parking management algorithm are determined, system 10 enumerates possible future railyard states based on evaluation of the initial railyard state and a yard schedule. Next system 10 examines each possible future railyard state wherein a cost and a time based efficiency of each possible future state is calculated.
Finally, the parking management algorithm optimizes costs and efficiency of each future state and determines an optimal present locomotive placement option. The optimal placement option is determined by comparing the costs of NSM's with the costs of delayed locomotive consist departure, by applying specific railyard locomotive management objectives and rules, and selecting the present placement option that will provide future states that the most closely fulfills the management objectives and rules. In one embodiment, management objectives include such things as, assembling an outbound train as scheduled, delivering the outbound train as scheduled, reducing labor involved in assembling and delivering the outbound train, and reducing delays in locomotive servicing. Thus, the parking decisions at any moment are based on an assessment of future state, utilizing specific criteria to sort through current parking options, both present and future, in order to assess the sum of immediate and future parking costs.
In an exemplary embodiment, management rules include parking yard management rules and service yard management rules. The parking yard management rules include such things as executing locomotive pull-forwards when there is a reduced number of locomotives on an affected track, maintaining an order of locomotives on each parking track such that locomotives for later outbound locomotive consists are parked behind locomotives for earlier outbound locomotive consists, and parking a lead locomotive for an outbound locomotive consist on a parking track such that the lead locomotive is in front of other locomotives parked on the same track that are allocated for the same outbound locomotive consist.
Locomotives due for service create a separate queuing problem, which is jointly handled with locomotive parking. As in the case of parking, the order in which locomotives are serviced affects the time at which they are available, and the general process of queuing them before and after service entails some inefficiencies. For example, a service shop may or may not have multiple bays, and the bays may or may not be served by separate lead-in tracks and separate tracks at the output of the shop. Thus, service yard management rules for making decisions as to order of service will be very specific to the service and parking facilities of a given yard. In an exemplary embodiment, the service yard management rules include such things as positioning a locomotive in a queue for service on a lead-in track to a service bay that provides the appropriate service, positioning locomotives in queue on a lead-in track in an order that allows servicing of each locomotive to be completed before each locomotive is scheduled for assembly in an outbound locomotive consist, and scheduling short service activities before long service activities when scheduling conflicts are not at issue. In other embodiments, other service yard management rules apply, based on the specifics of the service shop and railyard.
In order to determine an optimal present locomotive placement option, the locomotive parking management algorithm must evaluate each possible future parking configuration. In one embodiment, the algorithm applies a simple branching process, beginning with an enumeration of all possible present options, and then examining all possible future states, which might arise from each present option.
For example, if a railyard has four incoming locomotives and ten available parking slots there are,
N(1)=14!/(4!10!)=1001 possible parking arrangements.
If one of these parking arrangements is selected, and later two locomotives are assembled in an outbound train, and three more locomotives arrive with a second inbound train. Then at this time there will be,
N(2)=12!/(3!9!)=220 possible parking arrangements.
Furthermore, if a second outbound train departs, the next sequence of four incoming locomotives gives rise to,
N(3)=14!/(4!10!)=1001 possible parking arrangements.
Therefore, considering all possible parking arrangements for the first three inbound trains, the locomotive parking management algorithm must evaluate,
(1001)(1001)(220)=220,440,220 possible combinations.
There are many possible techniques that are applicable to calculate the number of possible future states. The branching process shown above is by way of example only, and is not intended to limit the possible techniques used by the locomotive parking management algorithm to evaluate future states.
Each workstation, 238, 240, and 242 is a personal computer having a web browser. Although the functions performed at the workstations typically are illustrated as being performed at respective workstations 238, 240, and 242, such functions can be performed at one of many personal computers coupled to LAN 236. Workstations 238, 240, and 242 are illustrated as being associated with separate functions only to facilitate an understanding of the different types of functions that can be performed by individuals having access to LAN 236.
In another embodiment, server system 212 is configured to be communicatively coupled to various individuals or employees 244 and to third parties, e.g., internal or external auditors, 246 via an ISP Internet connection 248. The communication in the exemplary embodiment is illustrated as being performed via the Internet, however, any other wide area network (WAN) type communication can be utilized in other embodiments, i.e., the systems and processes are not limited to being practiced via the Internet. In addition, and rather than a WAN 250, local area network 36 could be used in place of WAN 250.
In the exemplary embodiment, any authorized individual or an employee of the business entity having a workstation 254 can access the locomotive management system. One of the client systems includes a workstation 256 located at a remote location. Workstations 254 and 256 are personal computers having a web browser. Also, workstations 254 and 256 are configured to communicate with server system 212.
While the invention has been described in terms of various specific embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the claims.
This application claims the benefit of U.S. Provisional application No. 60/175,479, filed Jan. 11, 2000, which is hereby incorporated by reference in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
3861316 | Yamazaki et al. | Jan 1975 | A |
4610206 | Kubala et al. | Sep 1986 | A |
4630216 | Tyler et al. | Dec 1986 | A |
4926343 | Tsuruta et al. | May 1990 | A |
5623413 | Matheson et al. | Apr 1997 | A |
6311102 | Minakami | Oct 2001 | B1 |
6377877 | Doner | Apr 2002 | B1 |
6405127 | Doner | Jun 2002 | B1 |
6446912 | Barich et al. | Sep 2002 | B1 |
6456937 | Doner et al. | Sep 2002 | B1 |
20010044681 | Diana et al. | Nov 2001 | A1 |
Number | Date | Country |
---|---|---|
195 05 223 A 1 | Aug 1996 | DE |
0 721 872 | Jul 1996 | EP |
Number | Date | Country | |
---|---|---|---|
20010034642 A1 | Oct 2001 | US |
Number | Date | Country | |
---|---|---|---|
60175479 | Jan 2000 | US |