WO2016118122A1

WO2016118122A1 - Optimization of truck assignments in a mine using simulation

Info

Publication number: WO2016118122A1
Application number: PCT/US2015/012094
Authority: WO
Inventors: Kosta RISTOVSKI; Chetan Gupta; Kunihiko Harada; Hsiu-Khuern Tang
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2015-01-20
Filing date: 2015-01-20
Publication date: 2016-07-28
Anticipated expiration: 2017-07-20

Abstract

Example implementations described herein are directed to truck assignment solutions which can be utilized in the mining industry. Example implementations can be used to achieve improved asset utilization in all mines and is not constrained to any particular mine. Example implementations include conducting simulations to determine truck assignments to shovel/dump pairs.

Description

OPTIMIZATION OF TRUCK ASSIGNMENTS IN A MINE

USING SIMULATION

[0001] Field

[0002] The present application is generally directed to truck operations and more specifically, to truck distributions among shovel/dump pairs in a mining operation.

[0003] Background

[0004] In open pit mining, huge quantities of ore and waste material are transported using large equipment. The major components of material handling are trucks, shovels and loaders. Trucks, depending on the size and manufacturer, are organized into fleets. Depending on the material type, trucks may haul material from either shovels or loaders to the following destinations: dump areas/sites in the case of waste, and stockpiles or processing plants in the case of ore. Besides hauling, other main productive activities are material dumping, trucks driving empty, trucks loading, trucks spotting at shovel, and so on.

[0005] Due to the stochastic nature of activity durations and poor scheduling, nonproductive activities (NPT) such as truck queuing and shovel starving (e.g., waiting for trucks to be loaded) are present. In order to compete in the market and have sustainable and economical mining operations, companies attempt to improve their efficiency and reduce operational cost by decreasing the time spent in these non-productive activities. Truck assignment as a part of a dispatching system has a role to determine the number of trucks from each fleet that should be operating between any particular pair of loading (loaders, shovels) and dumping locations(dump areas) to meet production requirements. Material transportation can represent up to 40% of operating costs and hence reducing NPT in these systems can lead to savings for a mine operation.

Summary

[0006] Aspects of the present disclosure include a computer system, which may include a memory configured to store management information comprising topology information for a plurality of shovels and a plurality of dump sites, one or more truck parameters associated with a plurality of trucks, and one or more predicted activity durations for the plurality of trucks; and a processor, configured to process a specified shovel metric score associated with a confidence level, the confidence level determined from the one or more simulations; perform, based on the management information, one or more simulations, each of the one or more simulations associated with a truck distribution of one or more of the plurality of trucks among a pair comprising a shovel from the plurality of shovels and a dump site from the plurality of dump sites; calculate, for each of the one or more simulations, a shovel metric score for the plurality of shovels; and select, from the one or more simulations, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level, the associated truck distribution indicative of an assignment of trucks from the performing of the one or more simulations and a number of trucks needed from optimization.

[0007] Aspects of the present disclosure may further include a method, which can include managing management information comprising topology information for a plurality of shovels and a plurality of dump sites, one or more truck parameters associated with a plurality of trucks, and one or more predicted activity durations for the plurality of trucks; processing a specified shovel metric score associated with a confidence level, the confidence level determined from the one or more simulations; performing based on the management information, one or more simulations, each of the one or more simulations associated with a truck distribution of one or more of the plurality of trucks among a pair comprising a shovel from the plurality of shovels and a dump site from the plurality of dump sites; calculating for each of the one or more simulations, a shovel metric score for the plurality of shovels; and selecting from the one or more simulations, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level, the associated truck distribution indicative of an assignment of trucks from the performing of the one or more simulations and a number of trucks needed from optimization.

[0008] Aspects of the present disclosure may further include a computer program having instructions, which may include managing management information comprising topology information for a plurality of shovels and a plurality of dump sites, one or more truck parameters associated with a plurality of trucks, and one or more predicted activity durations for the plurality of trucks; processing a specified shovel metric score associated with a confidence level, the confidence level determined from the one or more simulations; performing based on the management information, one or more simulations, each of the one or more simulations associated with a truck distribution of one or more of the plurality of trucks among a pair comprising a shovel from the plurality of shovels and a dump site from the plurality of dump sites; calculating for each of the one or more simulations, a shovel metric score for the plurality of shovels; and selecting from the one or more simulations, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level, the associated truck distribution indicative of an assignment of trucks from the performing of the one or more simulations and a number of trucks needed from optimization. The instructions may be stored in a computer readable medium such as a non-transitory computer readable medium, wherein the instructions are executed by a processor.

Brief Description of Drawings

[0009] · FIG. 1 illustrates an example operation of trucks and shovels, in accordance with an example implementation.

[0010] · FIG. 2 illustrates a representation of a truck assignment problem, in accordance with an example implementation.

[0011] · FIG. 3 illustrates a logical view of a truck assignment system, in accordance with an example implementation.

[0012] · FIG. 4 illustrates a flow diagram overview of the truck assignment system, in accordance with an example implementation.

[0013] · FIG. 5 illustrates a hardware diagram of the truck assignment system, in accordance with an example implementation.

[0014] · FIG. 6 illustrates an example of truck information in accordance with an example implementation.

[0015] · FIG. 7 illustrates an example of topology information, in accordance with an example implementation.

[0016] · FIG. 8 illustrates an example of truck activity information, in accordance with an example implementation. [0017] · FIG. 9 illustrates an example of shovel/fleet match information, in accordance with an example implementation.

[0018] · FIG. 10 illustrates an example of deployment information, in accordance with an example implementation.

[0019] · FIG. 11 illustrates a flow diagram for conducting simulations, in accordance with an example implementation.

[0020] · FIG. 12 illustrates a flow diagram for managing a machine learning model, in accordance with an example implementation.

[0021] · FIG. 13 illustrates a flow diagram for managing simulations, in accordance with an example implementation.

[0022] · FIG. 14 illustrates a simulation table, in accordance with an example implementation.

Detailed Description

[0023] The following detailed description provides further details of the figures and example implementations of the present application. Reference numerals and descriptions of redundant elements between figures are omitted for clarity. Terms used throughout the description are provided as examples and are not intended to be limiting. For example, the use of the term "automatic" may involve fully automatic or semi-automatic implementations involving user or administrator control over certain aspects of the implementation, depending on the desired implementation of one of ordinary skill in the art practicing implementations of the present application. Truck assignment and truck distribution may be used interchangeably.

[0024] Example implementations described herein are directed to achieving a desired level of a shovel metric (e.g. utilization) with a given confidence. Additional constraints can include utilizing a minimum number of trucks, imposing fleet level constraints, imposing route level constraints and considering the stochastic nature of the problem.

[0025] In example implementations, the truck assignment can be viewed as a solution to a constrained stochastic optimization problem with an objective to reduce operating cost. The stochastic nature of the problem comes from the stochastic activity durations involved in mining field operations.

[0026] Example implementations may consider the following for providing a solution to the truck assignment problem. Example implementations may utilize machine learning models and historical data to predict the duration of activities and parameters of activity duration distribution. In addition, example implementations may involve computing an estimate for the minimum number of trucks under deterministic scenario such that a target shovel metric is obtained along with fleet and route constraints are met. The deterministic scenario uses expected values of activity durations obtained from machine learning models. This estimate is used as the total number of trucks in the following step. Further, example implementations may involve computing an assignment of a given total number of trucks between shovel and dump pairs in deterministic scenario while satisfying the fleet and route constraints. Thus, the stochastic part of the problem can be addressed such that the assignment is tested with one or more simulations, and if the desired confidence in the shovel metric is not achieved, the number of trucks is incremented by one and the simulation process is repeated. The activity duration distribution parameters can be used in the simulation.

[0027] Thus, example implementations address the problem of achieving a desired shovel metric (user supplied) with a given confidence level, the constraints, and the stochastic nature of the problem while minimizing the number of trucks. The problem can be modeled as a stochastic optimization problem, which is solved by combining methods from machine learning, convex integer optimization, and simulation.

[0028] Parameters of the distributions of activity durations, which are used in both the optimization procedure and the simulation, can be obtained as output of a machine learning model which takes into account several variables such as terrain, weather, type of truck, etc., instead of simple averages as utilized in related art implementations.

[0029] The stochastic optimization problem can be solved in three parts: the convex integer optimization is used to estimate the minimum number of trucks, the convex integer optimization is used to compute the assignment of trucks given the number of trucks, and simulation is used to address stochastic nature of the problem by computing the confidence in the given assignment. Example implementations of the simulator utilize the parameters of the distributions of activity durations computed by the machine learning models.

[0030] FIG. 1 illustrates an example operation of trucks and shovels, in accordance with an example implementation. The mining operation may include a plurality of shovels 101, a plurality of trucks 104, and dump sites 103. Trucks 104 and/or shovels 101 may be communicatively coupled to a computer system 102 through a network 100. Trucks 104 may navigate to shovels 101 to receive a payload and may also form a queue in front of shovels 101 when the shovels are being utilized. Trucks may also navigate to dump sites 103 to offload the payload.

[0031] Truck Assignment Problem

[0032] Example implementations generate the truck assignment in an attempt to improve efficiency and reduce cost of mine operations over some operating window, e.g. one shift, by maximizing an objective function and using historical data. Specifically, the objective of the truck assignment may be to find the truck assignment for given shovels, their locations and dump area locations, and to use a minimum number of trucks such that some shovel-related target metric is above a given threshold with a certain confidence. This means that the truck assignment can maximize production related objective by using minimum number of trucks.

[0033] FIG. 2 illustrates a representation of a truck assignment problem, in accordance with an example implementation. The truck assignment problem is shown where the general cases are considered as follows. The number of shovels used in the operations is s, number of dump areas is d, and trucks from different shovels can dump material at the same place. Trucks from multiple dumps can travel to the same shovel, and truck fleets, depending on characteristics, may not travel between certain shovel-dump pairs because of a shovel matching problem (e.g., fleet level constraint) or road constraints (e.g., route level constraint).

[0034] System for Truck Assignment

[0035] FIG. 3 illustrates a logical view of a truck assignment system, in accordance with an example implementation. Sensor data coming from the equipment 101, 104 will be processed through a complex event processing/streaming engine (CEP) 300 in real time where the trigger for the truck assignment can be generated if, for example, a shovel breaks down or a new shift is about to begin.

[0036] Data is processed by the computer system 102 and stored in a relational database 304. After the trigger is generated, machine learning models 303 may predict: (i) activity durations and (ii) distribution parameters of activity durations using historical data obtained from the database.

[0037] The outputs of the machine learning models as well as data from the database are used as input parameters for optimization modules 301. The outputs of both simulation 302 and machine learning 303 along with the data from database 304 are used in the stochastic optimization that may generate an optimized truck assignment. The obtained truck assignment can be displayed on a dashboard 305 so that a dispatcher 306 can deploy the distribution to the trucks and/or the shovels.

[0038] FIG. 4 illustrates a flow diagram overview of the truck assignment system, in accordance with an example implementation. At 400, the computer system 102 is configured to process a specified shovel metric, which can be provided by the dispatcher or automated depending on the desired implementation. The shovel metric may be related to shovel utilization, or other desired metrics as described herein. At 401, the computer system 102 executes one or more simulations based on the available truck fleet and dump site management information. At 402, the computer system selects a truck distribution based on the one or more simulations. At 403, the truck distribution is dispatched to the trucks and/or the shovels, whereupon the trucks and/or the shovels can adjust their schedules and shifts accordingly.

[0039] FIG. 5 illustrates a hardware diagram for a computer system, in accordance with an example implementation. Computer system 102 may be implemented as a management computer which is configured with a processor 501, memory 502, local disk 503, input/output (I/O) device 504 and local area network interface (LAN I/F) 505. Memory 502 may be implemented the form of a storage such as a storage system, a computer readable medium, random access memory (RAM) and so forth depending on the desired implementation. Memory 502 may be configured to store truck information 502-01, topology information 502-02, truck activity information 502-03, shovel/fleet match information 502-04, deployment information 502-05, a learning/simulation process 502- 05, a simulation table 502-10, and a mining operation database 502-11. Processor 501 may be configured to refer to memory 502 and invoke the learning/simulation process 502-06 as needed to implement the flow diagrams as described herein.

[0040] Data tables needed for our truck assignment solution

[0041] In example implementations, the machine learning model for activity durations are built to utilize as much relevant data as needed. Depending on activity, explanatory variables can be obtained from truck activity, topology, and truck details based on information stored in the memory of the computer system. Such variables can include shift information, weather data, route characteristics, truck health data such as original equipment manufacturer (OEM) data and so on. For the machine learning model to learn to predict each activity duration, the durations are provided in truck activity information 502-03. Constraints can be constructed from viable fleet- shovel combinations as well as from the topology information 502-02. Viable fleet-shovel combinations are managed in shovel/fleet match information 502-04 which contains information what truck fleet can be served by particular shovel. This allows for example implementations to avoid matching a big truck with a small shovels which otherwise would lead to inefficiency. The number of trucks per fleet can be calculated from the truck information 502-01. Further details are provided below.

[0042] FIG. 6 illustrates an example of truck information 502-01 in accordance with an example implementation. Truck information may include the truck identifier, the fleet identifier, and OEM information. Such OEM information can include the odometer reading, the truck model, maintenance time for a truck and payload capacity. Depending on the desired implementation, the truck information 502-01 may include other variables or omit any one of the listed variables.

[0043] FIG. 7 illustrates an example of topology information 502-2, in accordance with an example implementation. Topology information 502-02 may include shovel identifier, dump site identifier, distance between shovel and dump and route characteristics. Such route characteristics can include the elevation gradient for the route between the shovel and the corresponding dump site and route conditions (e.g., paved, mud, gravel, etc.). Depending on the desired implementation, the topology information 502-02 may include other variables or omit any one of the listed variables. [0044] FIG. 8 illustrates an example of truck activity information 502-03, in accordance with an example implementation. Truck activity information 502-03 can include the truck identifier/number, the shovel identifier/number, the dump site identifier/number, shift information, activity information, weather data (e.g., temperature, snow conditions, heavy wind, rain conditions etc.), and activity durations. Depending on the desired implementation, the truck activity information 502-03 may include other variables or omit any one of the listed variables.

[0045] FIG. 9 illustrates an example of shovel/fleet match information 502-04, in accordance with an example implementation. Shovel/Fleet match information 502-04 can include fleet number/identifier and shovel number/identifier and is indicative as to what fleets can travel to which shovels.

[0046] FIG. 10 illustrates an example of deployment information 502-05, in accordance with an example implementation. Deployment information 502-05 is indicative of a truck distribution for a fleet, and associates trucks to shovel/dump pairs. Deployment information 502-05 can include truck identifiers, shovel identifiers, and dump site identifiers. Each simulation can be associated with a corresponding deployment information 502-05.

[0047] Solution for the Truck Assignment Problem

[0048] Example implementations define the optimization problem as

subject to

additional constrainti additional_constraint2 additional constraint -,m

[0049] where f as the total number of truck fleets, x(S_itD_j,F is the number of trucks from fleet F_k that are traveling between shovel Si and dump area Dp M(Si) is shovel-related target metric such as utilization, tonnage, etc., c; is a pre-specified constant which equals to the target value of a given metric M(Sj), N(F_k) is the number of trucks in the fleet, and Jl is a specified subset of unviable triplets (S DpF_k), possibly empty. The other expressions are as follows:

[0050] · The objective function may be directed to minimizing the total number of trucks over all fleets.

[0051] · The first set of constraints imposes that the achieved metrics for shovels are equal or greater than target value. In example implementations it is possible to assign a different metric to each shovel. Depending on the desired implementation, replacing these constraints with one aggregate constraint (e.g., a sum over all shovels), may represent a system level constraint instead of having the set of constraints for individual shovels.

[0052] · The second set of constraints may include a prohibition from using more trucks than allowed in a particular fleet.

[0053] · The third set of constraints may ensure that each route has non-negative number of trucks.

[0054] · The fourth set of constraints may impose fleet and route level constraints.

[0055] · Moreover, additional constraints can be added to customize optimization problem when the problem remains convex.

[0056] For example, for the first set of constraints, in the case of using utilization of shovels as shovel related metric, it is defined over time period T as

[0057] where t _e (Si) is idle time of shovel Si. The idle time of the shovels is function of activity times during the operations and is therefore a stochastic variable as well as utilization of shovels.

[0058] Because of stochasticity, the first set of constraints can be satisfied only under some pre-specified confidence level. The approach to solving this stochastic optimization problem while meeting the confidence level in shovel utilization is described in further detail in FIG. 1 1.

[0059] FIG. 1 1 illustrates a flow diagram for conducting simulations, in accordance with an example implementation. At 1 100, the target value of the shovel metric and a required confidence level is specified. The shovel metric and required confidence can be provided by a user through a user interface, derived based on the mining operations or by other methods depending on the desired implementations. At 1 101 , the computer system reads the location of the shovels, dump areas, and available trucks from the memory, such as the information as illustrated in FIGS. 6-10. Depending on the desired implementation, the computer system may also read historical information from the mining operation as well as other mining operations for the construction of machine learning models.

[0060] At 1 102, the computer system develops machine learning models configured to predict parameters of non-negative distributions for activity durations. The machine learning models can then be implemented in the simulations. At 1 103, the computer system calculates an initial guess for the minimum number of trucks. The initial guess can be conducted in accordance with the desired implementation. For example, the initial guess can be based on a convex integer optimization in a deterministic scenario or by other methods.

[0061] The process at 1 104-1 107 illustrates an iterative process for conducting one or more simulations based on the initial guess. At 1 104, trucks are assigned to shovel/dump pairs and their corresponding routes by the computer system. The assignment can be conducted, for example, by using convex integer optimization in the deterministic scenario as described above. However, other implementations are also possible, depending on the desired implementation. At 1 105, a simulation model is run by using the truck assignment and the predicted activity durations to obtain the confidence level for the shovel metric by the computer system. Depending on the desired implementation, each simulation may be executed multiple times to obtain the confidence level and to provide an average or other aggregated calculation of the shovel metric across the multiple executions. At 1106, a determination is made by the computer system to determine if the confidence level is satisfied. If so (Yes) then the resulting truck distribution is utilized for dispatching trucks in the mining operation. If not (no), then the simulation proceeds to reiterate the simulation after incrementing the total number of trucks in the initial guess utilized in the simulation as shown at 1107.

[0062] Any metric can be used for the shovel metric, depending on the desired implementation. For example, if tonnage is used as a metric instead of shovel utilization, the tonnage metric can be calculated as a utilization of shovels multiplied by the loading rate of the shovel. Then the optimization problem can be similarly implemented.

[0063] Machine learning models

[0064] Example implementations may utilize machine learning models to predict the future more accurately than simple averages over historical data. To predict the future, machine learning takes in explanatory variables as input. In this case, examples of explanatory variables which can be used for prediction of distribution parameters of activity durations are following:

[0065] Distance between loading locations and dump areas,

[0066] Route profile (e.g. positive and negative elevation gradients on the route),

[0067] Shift,

[0068] Fleet,

[0069] Weather data.

[0070] These explanatory variables can be known in the near future from road information as well as weather forecast. Shift is an explanatory variable because conditions may be different during the night than during the day. Weather data may also be an explanatory variable, since in the case of rain, snow or high winds, the trucks may move slower than usual. Machine learning models can be developed for each of the activities separately. Each of machine learning models can be configured to be able to model non-negative distribution (Gamma, Weibull, Log-normal) because time as a response variable is non- negative. Machine learning models that can be deployed in this solution include but are not limited to: generalized linear models, neural networks, and hidden Markov models (HMMs). Parameters of these machine learning models can be learned using historical data. Once parameters are learned, these models are ready to be applied on new coming data by using the same parameters. Historical data and new coming data has to be in the same format in order to apply machine learning model. An example of the flow diagram is shown in FIG. 12.

[0071] FIG. 12 illustrates a flow diagram for managing a machine learning model, in accordance with an example implementation. The flow diagram corresponds to the flow at 1102 from FIG. 11. At 1200, data is applied from previous mining operations to a machine learning model. At 1201, the machine learning model is generated from the application of the data to a model building process as described above. At 1202, predictions are generated based on the execution of the machine learning model used in the simulation. The predictions are compared to the results of the mining operation wherein the machine learning model is updated accordingly as illustrated at 1203.

[0072] Mine Simulator

[0073] A mine simulator is built to simulate mining operations over time.

[0074] Because of inherent stochasticity in activity times there is queuing in the system and the simulator is used to address stochasticity and queuing. The simulation is configured to be capable of modeling the queuing effect which may not be possible to estimate by machine learning. The simulator incorporates the stochasticity of activity durations by sampling them from provided distributions. For improving accuracy and the modeling of the queuing effect, the mine simulator is supported with parameters of the distributions of activity times obtained from machine learning models. For example, the convex integer optimization can provide the truck assignments to the simulation as described above.

[0075] Initial guess

[0076] To avoid searching over all possible number of trucks that can be used in the mine as well as all possible truck assignments, an initial guess for minimum number of trucks is computed as shown at 1103 of FIG. 11. This initial guess can be calculated based on the convex integer optimization which involves expected values of activity times and neglects the queuing effect.

[0077] Therefore, in this case the first constraint from the optimization problem defined above becomes

[0078] For example, to completely formulate the optimization problem in the case of shovel utilization as a shovel metric, example implementations can define expected values of shovel idle time over time period T as

[0079] where t_cyci_e (Si,D_j,Ft) is time of the hauling cycle for truck from fleet F_k between shovel Si and dump D_j while ti_oad (Si,F_k) is time needed for shovel Si to load truck from fleet F_k. The hauling cycle is defined as a set of truck activities from the time point when truck leaves a dump to the time point when it finishes dumping. The expected value of the haul cycle time is defined as a sum of expected values of each of activity durations

where t_empty (Si,D_j,F_k), t_spot (F_k), ti_oad {S F_k), haul {S D_j,F_k) and t_dump (D_j,F_k) are activity durations for driving empty truck, truck spotting, truck loading, truck hauling material and truck dumping material respectively. Then, the optimization problem to obtain initial guess becomes

subject to

additional constrainti additional_constraint2 additional constraint_m

[0080] The problem can be solved by using integer programming, or other methods depending on the desired implementation.

[0081] Assignment of trucks to routes

[0082] For a given total number of trucks, the assignments of trucks to shovel-dump pairs is computed without necessarily exploring all of the possible combinations of assignments of trucks to shovel-dump pairs. Therefore a convex integer optimization can be utilized which will find the truck assignment. The optimization can provide a solution given the total number of trucks such that sum of expected values of given metrics across all shovels is maximized as well as expected metric value for each shovel is above the threshold. The optimization is defined as: m &xl misse ^ w^E [M_t

i-i

subject to

1,„„., . 8

£[¾;(¾>] < q + p; f = l,_^.., _fs

Y x{% ¾}≤ C¾) ; k = l,

1=1 1=1. additional constrainti additional_constraint₂ additional constraint -,m

[0083] where is weight to adjust for scales if different metrics are used, X is the total number of trucks to be assigned, and p is non-negative constant which controls over- trucking.

[0084] The second constraint should be utilized as otherwise the truck assignment solution can make some of the routes over-trucked which will increase truck queuing. Over- trucking is possible in absence of the second constraint because the optimization over expected values is not aware of the queuing effect.

[0085] The third constraint imposes that total number of trucks assigned to routes does not exceed the given number of trucks. In the case of unrealistic requirements set by the user, it is possible that there is no solution that can be produced which satisfies the requirements. For example, it is not feasible to have 100% utilization of shovels with 100% confidence. In this case the mathematical solution is an infinite number of trucks. In the case of unrealistic requirements, example implementations can be configured to return the last feasible solution obtained by re-assignment optimization to the dispatcher.

[0086] Below is the example of the above optimization function when shovel utilization is used as a metric:

subject to

q -f- p; i = -U~._*s

additional constrainti additional_constraint2 additional constraint -,m

[0087] FIG. 13 illustrates a flow diagram for managing simulations, in accordance with an example implementation. In example implementations, the computer system may receive feedback from the mining operations and compare the actual results of the metric compared to the target metric. If the actual results do not meet the target metric (e.g., due to change in conditions such as weather, available trucks, etc.) within a threshold, then the machine learning model may be updated with the actual results and more simulations are conducted to determine a new truck assignment. The new truck assignments are then presented to the user interface for selection.

[0088] At 1301, the computer system selects the truck distribution from the simulations as illustrated in FIG. 11 and as further described in FIG. 14. At 1302, the computer system updates the simulation table based on feedback of mining operations and new simulations. At 1303, the additional truck assignments are provided for selection, which can be implemented in a user interface.

[0089] FIG. 14 illustrates a simulation table 502-10, in accordance with an example implementation. The simulation table can include one or more truck assignments associated with the corresponding simulation which are indexed based on the simulation. Each simulation can be associated with deployment information as illustrated in FIG. 10. The simulation table can include a simulation identifier, the corresponding deployment information, the metric score yielded by the simulation and the confidence level. Depending on the desired implementation, the simulation table 502-10 may include or omit any of the listed information when presented to the user interface.

[0090] Finally, some portions of the detailed description are presented in terms of algorithms and symbolic representations of operations within a computer. These algorithmic descriptions and symbolic representations are the means used by those skilled in the data processing arts to convey the essence of their innovations to others skilled in the art. An algorithm is a series of defined steps leading to a desired end state or result. In example implementations, the steps carried out require physical manipulations of tangible quantities for achieving a tangible result.

[0091] Unless specifically stated otherwise, as apparent from the discussion, it is appreciated that throughout the description, discussions utilizing terms such as "processing," "computing," "calculating," "determining," "displaying," or the like, can include the actions and processes of a computer system or other information processing device that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system's memories or registers or other information storage, transmission or display devices.

[0092] Example implementations may also relate to an apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, or it may include one or more general-purpose computers selectively activated or reconfigured by one or more computer programs. Such computer programs may be stored in a computer readable medium, such as a computer-readable storage medium or a computer-readable signal medium. A computer-readable storage medium may involve tangible mediums such as, but not limited to optical disks, magnetic disks, read-only memories, random access memories, solid state devices and drives, or any other types of tangible or non-transitory media suitable for storing electronic information. A computer readable signal medium may include mediums such as carrier waves. The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Computer programs can involve pure software implementations that involve instructions that perform the operations of the desired implementation. [0093] Various general-purpose systems may be used with programs and modules in accordance with the examples herein, or it may prove convenient to construct a more specialized apparatus to perform desired method steps. In addition, the example implementations are not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the example implementations as described herein. The instructions of the programming language(s) may be executed by one or more processing devices, e.g., central processing units (CPUs), processors, or controllers.

[0094] As is known in the art, the operations described above can be performed by hardware, software, or some combination of software and hardware. Various aspects of the example implementations may be implemented using circuits and logic devices (hardware), while other aspects may be implemented using instructions stored on a machine-readable medium (software), which if executed by a processor, would cause the processor to perform a method to carry out implementations of the present application. Further, some example implementations of the present application may be performed solely in hardware, whereas other example implementations may be performed solely in software. Moreover, the various functions described can be performed in a single unit, or can be spread across a number of components in any number of ways. When performed by software, the methods may be executed by a processor, such as a general purpose computer, based on instructions stored on a computer-readable medium. If desired, the instructions can be stored on the medium in a compressed and/or encrypted format.

[0095] Moreover, other implementations of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the teachings of the present application. Various aspects and/or components of the described example implementations may be used singly or in any combination. It is intended that the specification and example implementations be considered as examples only, with the true scope and spirit of the present application being indicated by the following claims.

Claims

CLAIMS What is claimed is:

1. A computer system, comprising: a memory configured to store management information comprising topology information for a plurality of shovels and a plurality of dump sites, one or more truck parameters associated with a plurality of trucks, and one or more predicted activity durations for the plurality of trucks; a processor, configured to: process a specified shovel metric score associated with a confidence level, the confidence level determined from one or more simulations; perform, based on the management information, the one or more simulations, each of the one or more simulations associated with a truck distribution of one or more of the plurality of trucks among a pair comprising a shovel from the plurality of shovels and a dump site from the plurality of dump sites; calculate, for each of the one or more simulations, a shovel metric score for the plurality of shovels; and select, from the one or more simulations, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level, the associated truck distribution indicative of an assignment of trucks from the performing of the one or more simulations and a number of trucks needed from optimization.

2. The computer system of claim 1, wherein the processor is further configured to calculate, for the one or more simulations, the predicted activity durations based on machine learning.

3. The computer system of claim 1, wherein the associated truck distribution is a subset of the plurality of the trucks.

4. The computer system of claim 1, wherein the processor is configured to perform the one or more simulations by: providing an initial guess using optimization for the associated truck distribution for a first one of the one or more simulations; computing the assignment of trucks indicated in the initial guess by using optimization; and adding additional ones of the plurality of trucks to the truck distribution and computing the assignment for subsequent ones of the one or more simulations until the associated truck distribution has the shovel metric score that satisfies the specified shovel metric score within the confidence level.

5. The computer system of claim 1, wherein the specified shovel metric score is based on shovel utilization.

6. The computer system of claim 1, wherein each of the one or more simulations is executed multiple times, and wherein the shovel metric score is an average of the multiple executions.

7. The computer system of claim 1, wherein the computer system is communicatively coupled to a truck distribution system, the truck distribution system configured to communicate the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level to the plurality of trucks.

8. A method, comprising: managing management information comprising topology information for a plurality of shovels and a plurality of dump sites, one or more truck parameters associated with a plurality of trucks, and one or more predicted activity durations for the plurality of trucks; processing a specified shovel metric score associated with a confidence level, the confidence level determined from one or more simulations; performing based on the management information, the one or more simulations, each of the one or more simulations associated with a truck distribution of one or more of the plurality of trucks among a pair comprising a shovel from the plurality of shovels and a dump site from the plurality of dump sites; calculating for each of the one or more simulations, a shovel metric score for the plurality of shovels; and selecting from the one or more simulations, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level, the associated truck distribution indicative of an assignment of trucks from the performing of the one or more simulations and a number of trucks needed from optimization.

9. The method of claim 8, further comprising calculating, for the one or more simulations, the predicted activity durations based on machine learning.

10. The method of claim 8, wherein the associated truck distribution is a subset of the plurality of the trucks.

11. The method of claim 8, wherein the performing the one or more simulations comprises: providing an initial guess using optimization for the associated truck distribution for a first one of the one or more simulations; computing the assignment of trucks indicated in the initial guess by using optimization; and adding additional ones of the plurality of trucks to the truck distribution and computing the assignment for subsequent ones of the one or more simulations until the associated truck distribution has the shovel metric score that satisfies the specified shovel metric score within the confidence level.

12. The method of claim 8, wherein the specified shovel metric score is based on shovel utilization.

13. The method of claim 8, wherein each of the one or more simulations is executed multiple times, and wherein the shovel metric score is an average of the multiple executions.

14. The method of claim 8, further comprising communicating, through a truck distribution system, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level to the plurality of trucks.

15. A computer program having instructions, comprising: managing management information comprising topology information for a plurality of shovels and a plurality of dump sites, one or more truck parameters associated with a plurality of trucks, and one or more predicted activity durations for the plurality of trucks; processing a specified shovel metric score associated with a confidence level, the confidence level determined from one or more simulations; performing based on the management information, one or more simulations, each of the one or more simulations associated with a truck distribution of one or more of the plurality of trucks among a pair comprising a shovel from the plurality of shovels and a dump site from the plurality of dump sites; calculating for each of the one or more simulations, a shovel metric score for the plurality of shovels; and selecting from the one or more simulations, the associated truck distribution having the shovel metric score that satisfies the specified shovel metric score within the confidence level, the associated truck distribution indicative of an assignment of trucks from the performing of the one or more simulations and a number of trucks needed from optimization.