KR102469971B1 - Optimized maintenance schedule calculation method and apparatus for railway system - Google Patents

Abstract

The present invention relates to an optimized maintenance schedule calculation method for a railway system component, in a maintenance schedule calculation method for calculating a maintenance schedule of a component constituting a railway system, which comprises the steps of: collecting diagnosis score data for the state of a railway system component; calculating a state diagnosis prediction value of a railway system component from the score data; and calculating a maintenance schedule of the railway system component from the state diagnosis prediction value.

Description

Optimized maintenance schedule calculation method and device for railway system parts

The present invention relates to techniques for calculating maintenance schedules for railway system components.

Among the railway system components, there are those that are directly related to railway operation. One of them may be a tire used for light rail or the like. When a problem occurs in a tire, it is brought to the maintenance base, and a worker visually checks the problem and replaces it.

The tire replacement time is approximately 2 hours, and a separate maintenance schedule is established to replace the tire. At this time, in order to maximize the availability rate of the railway vehicle, not only the problem tire is replaced, but also the tire in good condition is replaced together. In addition, there is a possibility that work on railway vehicles to be stored and maintained at the maintenance base may occur at a certain level or more during the same period.

Accordingly, the inventors of the present invention have studied for a long time on a method for maximizing the availability rate of railroad vehicles during the planning period through the establishment of an optimized maintenance schedule for railroad system parts that are very closely related to railroad operation, such as tires, and after trial and error, The present invention has been completed.

An object of the present invention is to provide a method and apparatus for calculating an optimized maintenance schedule for railway system components capable of maximizing the availability rate of railway vehicles during a planning period.

In addition, it is intended to provide a maintenance schedule calculation method and apparatus capable of reducing costs by allowing parts to be replaced only within a certain period before reaching a threshold value.

Meanwhile, other unspecified objects of the present invention will be additionally considered within the scope that can be easily inferred from the following detailed description and effects thereof.

In the maintenance schedule calculation method for calculating the maintenance schedule of parts constituting the railway system according to an embodiment of the present invention,

collecting diagnostic score data on the condition of rail system components;

Calculating a condition diagnosis prediction value of a railway system component from the score data;

Characterized in that it comprises the step of calculating the maintenance schedule of the railway system parts from the condition diagnosis prediction value,

Provides an optimized maintenance schedule calculation method for railway system parts.

Calculating the predicted value according to an embodiment of the present invention,

It may include calculating a condition diagnosis predicted value from the score data through an ARIMA (AutoRegressive Integrated Moving Average) model.

Establishing the ARIMA model according to an embodiment of the present invention,

confirming the normality of the score data - calculating normal data by difference in case of abnormal data;

Defining and initial setting a prediction period and minAIC (Akaike Information Criterion);

Deriving time series data according to an autoregressive order;

Estimating ARMA model parameters (p, q) through Maximum Likelihood Estimation (MLE) method;

는 예측 오차의 분산, N은 모형 수립 시 데이터 수임.Calculating AIC (p, q), which is the goodness of fit of the model - the AIC (p, q) is defined by the formula below, where

is the variance of the prediction error, and N is the number of data when establishing the model.

-;

-;

If the AIC(p, q) is less than minAIC and the degrees p and q are greater than or equal to log(N), calculating a state diagnosis predicted value of the next step 1; and

It may include repeating the step of calculating the condition diagnosis prediction value from the step of arranging the prediction period into the time series up to the defined prediction period by increasing the prediction periods p and q by one step.

Calculating the maintenance schedule according to an embodiment of the present invention,

The maintenance schedule can be calculated by considering all of the minimization of unavailability of railway vehicles, the minimization of the residual condition diagnosis prediction value before maintenance, and the minimization of maintenance work occurring during the same period.

According to an embodiment of the present invention, the objective function for calculating the maintenance schedule may be defined by the following equation.

MIN

At this time,

i: index for rolling stock formation, i=1, 2, … I

j: Index of tires for each train group, j = 1, 2, . . . J

k: Index of the number of unavailable trains, k=0, 1, 2, … I

(Maximum cost of ownership = total number of trains)

n: Index of the number of action tasks, n=0, 1, 2, … N

t : index for period, t=1, 2, … T

P ¹ _k : Penalty weight according to the number of unavailable formations

D _kt : variable that determines the number of unavailable trains in period t

(D _kt =1 if k flight unavailable, otherwise 0)

^P2 : Weight

R _ijt : Remaining condition score before action for railroad car i organization j tire in period t

P ³ _n: Weight

Y _nt : means a variable that determines the number of action tasks in period t (Y _nt =1 for n tasks, otherwise 0).

According to an embodiment of the present invention, a data collection unit for collecting diagnostic score data for component states in order to predict the state of railroad system components; diagnosing the state of railroad system components for a certain period of time by applying an ARIMA model from the score data a state diagnosis prediction value calculation unit that predicts; and

Characterized in that it comprises a maintenance schedule calculation unit for calculating the maintenance schedule of railway system parts through an objective function from the condition diagnosis predicted value,

It is possible to provide an optimized maintenance schedule calculator for railway system parts.

Using the optimized maintenance schedule calculation method and device for railway system components according to the present invention, it is possible to maximize the availability rate of railway vehicles during the planning period.

In addition, it is possible to obtain an effect of cost reduction by allowing parts to be replaced only within a certain period before reaching the threshold value.

On the other hand, even if the effects are not explicitly mentioned here, it is added that the effects described in the following specification expected by the technical features of the present invention and their provisional effects are treated as described in the specification of the present invention.

1 is a block diagram of a device for calculating an optimized maintenance schedule for railway system components according to the present invention.
2 is a flowchart illustrating a method for calculating an optimized maintenance schedule according to the present invention.
3 is a flowchart illustrating the calculation of condition diagnosis prediction values according to the ARIMA model.
4 is a diagram showing predicted values calculated according to the ARIMA model for each part.
5 is a diagram showing a maintenance schedule for each train group.
It is revealed that the accompanying drawings are illustrated as references for understanding the technical idea of the present invention, and thereby the scope of the present invention is not limited thereto.

In the description of the present invention, if it is determined that a related known function may unnecessarily obscure the subject matter of the present invention as an obvious matter to those skilled in the art, the detailed description thereof will be omitted.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that it does not preclude the possibility of the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

Hereinafter, a method for calculating an optimized maintenance schedule for railway system parts according to the present invention and an embodiment of the device will be described in detail with reference to the accompanying drawings, and in the description with reference to the accompanying drawings, the same or corresponding components are given the same reference numerals, and overlapping descriptions thereof will be omitted.

The present invention relates to a method for calculating an optimized maintenance schedule for railway system parts, and largely maximizes the availability rate of railway vehicles during a certain planning period, while replacing parts only within a limited period before the threshold of parts occurs. It is to obtain the effect of cost reduction by enabling maintenance of

In addition, it is possible to manage not to exceed the processing capacity that can be processed at the maintenance base by preventing action work from occurring at a level higher than a certain level in the same period.

To this end, the present invention is characterized in that a predicted value of a condition diagnosis of a railway system component is calculated, and a maintenance schedule is calculated by introducing various constraints therefrom.

1 is a block diagram of a device for calculating an optimized maintenance schedule for railway system components according to the present invention.

As shown in FIG. 1, the optimized maintenance schedule calculator (100, hereinafter referred to as 'maintenance schedule calculator') of the railway system parts according to the present invention includes a data collection unit 110, a condition diagnosis predicted value calculator ( 120), and a maintenance schedule calculator 130.

The data collection unit 110 collects diagnostic score data for component states in order to predict the state of railroad system components. Specifically, it is to collect score values for a specific diagnosis of the condition of parts requiring maintenance related to railway operation among railway system parts. For example, the condition of the health of railway system parts may be diagnosed, and the result may be digitized and input into a data collection unit as a score.

The score data collected in this way can be used as data for diagnosing the condition of railway system components for a certain period of time. The score data collected by the data collection unit 110 may include both normal data and abnormal data.

Accordingly, in the present invention, the condition diagnosis prediction value calculation unit 120 predicts the condition diagnosis of railway system parts for a certain period through the ARIMA model.

Thereafter, the maintenance schedule calculation unit 130 calculates the maintenance schedule of railway system components using the objective function. At this time, the objective function is designed in the direction of minimizing the sum of the penalty for non-availability by planning period, the sum of the remaining condition points before maintenance, and the sum of the penalty for actions that occur over a certain number of tasks during the same period.

2 is a flowchart illustrating a method for calculating an optimized maintenance schedule according to the present invention.

As shown in FIG. 2 , the method for calculating an optimized maintenance schedule for railway system components according to the present invention includes a diagnostic score data collection step, a condition diagnosis prediction value calculation step, and a maintenance schedule calculation step.

The diagnostic score data collection step is to collect diagnostic score data on the state of railway system parts, which are condition diagnosis scores for railway system parts obtained through actual measurement.

As an embodiment according to the present invention, the table below shows the condition diagnosis results for the soundness of light rail tires. As for the condition diagnosis score of the tire, one condition diagnosis score per day can be measured and collected.

The tire condition diagnosis score can be defined as a value between 1 and 100 points (100 = best, 1 = worst).

date organization chaho order tire id condition diagnosis score 2021-11-30 One 4001 One 4001-1 61 2021-11-30 One 4001 2 4001-2 59 2021-11-30 One 4001 3 4001-3 65 2021-11-30 One 4001 4 4001-4 63 2021-11-30 One 4101 One 4101-1 79 2021-11-30 One 4101 2 4101-2 80 2021-11-30 One 4101 3 4101-3 81 2021-11-30 One 4101 4 4101-4 80

In addition, the condition diagnosis score can be organized as follows in a time series by ID of a specific tire.

date organization chaho order tire id condition diagnosis score 2021-11-30 One 4001 One 4001-1 61 2021-11-29 One 4001 One 4001-1 59 2021-11-28 One 4001 One 4001-1 61 2021-11-27 One 4001 One 4001-1 60 2021-11-26 One 4101 One 4001-1 59 2021-11-25 One 4101 One 4001-1 64 2021-11-24 One 4101 One 4001-1 63 2021-11-23 One 4101 One 4001-1 66

These data are used to calculate the condition diagnosis prediction value of railway system parts. At this time, in the step of calculating the predicted value, the condition diagnosis predicted value is calculated from the condition diagnosis score data through the ARIMA (AutoRegressive Integrated Moving Average) model.

If these condition score data are normal data, they will follow the general form of the Autoregressive Moving Average (ARMA)(p, q) model.

The general form of the ARMA model is as follows.

는 AR의 모수,

는 MA의 모수,

는 정상 데이터이고

은 백색잡음(White Noise)를 나타낸다. Here, p is the order of Autoregressive (AR), q is the order of Moving Average (MA),

is the parameter of AR,

is the parameter of MA,

is the normal data

represents white noise.

KPSS test is used to confirm the stationarity of time series data.

The KPSS (Kwiatkowski-Phillips-Schmidt-Shin Test) test is one of the Unit-root test methods to test the stationarity of time series data. )" and, conversely, the alternative hypothesis is set as "the time series is non stationary".

Data stationarity is confirmed through the KPSS test, and differencing data is calculated for data that does not represent stationarity.

If abnormal data is included, it is differentiated and calculated as normal data, and then the condition diagnosis prediction value is calculated according to the ARIMA model.

는 정상 아닌 데이터, d는 차분 차수를 나타낸다.here

is non-normal data, and d represents the difference order.

3 is a flowchart illustrating the calculation of condition diagnosis prediction values according to the ARIMA model.

As shown in FIG. 3, calculating the condition diagnostic prediction value according to the present invention includes establishing an ARIMA model.

For the application of the ARIMA model, the normality of the score data is confirmed. In the case of abnormal data, normal data is calculated by difference. Thereafter, a step of defining a prediction period and minAIC (Akaike Information Criterion) and initial setting is performed.

Then, time series data is derived according to the autoregressive order, and the ARMA model parameters (p, q) are estimated through the MLE (Maximum Likelihood Estimation) method.

는 예측 오차의 분산, N은 모형 수립 시 데이터 수를 나타낸다.Then, AIC(p, q), which is the goodness of fit of the model, is calculated. At this time, AIC(p, q) is defined by the equation below, where

is the variance of the prediction error, and N is the number of data when establishing the model.

When the calculated AIC(p, q) is smaller than minAIC and the order p and q are greater than log(N), the next step 1 state diagnosis prediction value is calculated.

Thereafter, the step of arranging the prediction periods, p, and q in a time series while raising the predicted periods by one step until the defined prediction period, and the step of calculating the condition diagnosis prediction value are repeatedly performed.

4 is a diagram showing predicted values calculated according to the ARIMA model for each part.

As shown in FIG. 4, the condition diagnosis prediction value for each tire predicted through the 52-day ARIMA model is shown. That is, by calculating a predicted value for the future state of each tire, it is possible to calculate a tire maintenance plan based on this.

In the step of calculating the maintenance schedule, the maintenance schedule is calculated by considering all of the minimization of non-availability of railway vehicles, the minimization of the residual condition diagnosis prediction value before maintenance, and the minimization of maintenance work occurring during the same period.

Conventionally, in order to maximize the availability of railway vehicles, there is a possibility of replacing or maintaining tires even when the tires are in good condition. In addition, there is a possibility that maintenance or replacement work may occur simultaneously and intensively in the same period.

Accordingly, it is considered that measures such as maintenance are possible only within a limited period prior to the estimated threshold of each tire, and that the remaining state score before maintenance of each tire is minimized. In addition, maintenance work can be limited so that it does not occur more than a certain level in the same period so that it is not concentrated in a certain period.

Reflecting these points, an objective function for establishing a maintenance schedule was derived.

At this time, the objective function was made under the following assumptions.

1) The current condition score for each tire of each railway vehicle is known.

2) During the planning period, the condition score for each tire of each railroad car is predictable and definitive.

3) The minimum usable condition score (threshold) of a tire is known, and it is not allowed to generate unusable condition scores for all tires during the planning period. (This means that maintenance is required in the previous period based on the threshold occurrence time for each tire.)

4) The maintenance work type for each tire is single, and additional maintenance is not required during the planning period as it is restored to the highest condition score after maintenance work.

5) Maintenance for each tire is completed in a single period.

6) During the maintenance period of each tire, the railroad car equipped with the tire becomes unavailable.

7) Do not consider other maintenance schedules of rolling stock.

Under the above assumptions, the objective function for establishing the maintenance schedule was derived by considering the minimization of unavailability of railway vehicles, the minimization of the residual condition diagnosis prediction value before maintenance, and the minimization of maintenance work occurring during the same period.

The resulting objective function is as follows.

MIN

In other words, the maintenance plan is set so that the sum of the penalty for non-availability of railway vehicles, the estimated value of residual condition diagnosis before maintenance, and the penalty for maintenance work that occurs more than a certain number of work during the same period is minimized.

In the above-described objective function, the index, input parameters, and decision-making variables are as follows.

i: index for rolling stock formation, i=1, 2, … I

j: Index of tires for each train group, j = 1, 2, . . . J

k: Index of the number of unavailable trains, k=0, 1, 2, … I

(Maximum cost of ownership = total number of trains)

n: Index of the number of action tasks, n=0, 1, 2, … N

t : index for period, t=1, 2, … T

P ¹ _k : Penalty weight according to the number of unavailable formations

D _kt : variable that determines the number of unavailable trains in period t

(D _kt =1 if k flight unavailable, otherwise 0)

^P2 : Weight

R _ijt : Remaining condition score before action for railroad car i organization j tire in period t

P ³ _n: Weight

Y _nt : A variable that determines the number of action tasks in period t (Y _nt =1 for n tasks, otherwise 0)

5 is a diagram showing a maintenance schedule for each train group. As a maintenance schedule planning screen calculated according to the above-described objective function, only the calculated number of maintenance work and the number of unavailable trains are disclosed on this screen, but in more detail, specific information on the tire requiring maintenance is further provided. can include

As such, the present invention calculates the optimal maintenance schedule plan for each train organization by considering all of the minimization of the unavailability of railroad vehicles, the minimization of the residual state diagnosis prediction value before maintenance, and the minimization of maintenance work occurring during the same period, so that during the planning period While maximizing the availability of rail vehicles, it is also possible to reduce costs on the one hand.

In addition, constraints according to various assumptions and purpose definitions can be applied to the schedule plan for optimizing the tire maintenance of railway vehicles.

1) In the case of not allowing unusable state points to occur during the planning period, the following constraints can be applied.

At this time, S _min means an unusable state score (threshold).

The above-mentioned constraint expression means that the condition points of all tires must be greater than the minimum usable condition score (threshold) through necessary measures during the planning period, and there are maximum I × J × T number of constraint expressions.

2) The type of action work for each tire is single, and if additional action is not required during the planning period after the action is restored to the highest condition score, the following constraints can be applied.

The above-mentioned constraint formula restricts to consider only one action or less during the planning period, and there are maximum I × J constraint formulas.

At this time, NS _ijt means the estimated condition score for railroad car i formation j tire in period t when no action is taken, _Stop means the highest condition score that does not require additional action after action, and X _ijt means period t means whether or not there is an action (replacement 1, otherwise 0) for railway vehicle i organization j tires.

In the above-mentioned constraint formula, if an action on the tire of train i organization j occurs in the period t and the previous period, the condition score of the tire changes to _Stop , which is the highest condition score that does not require additional action, and if there is no action, the first estimated condition score It means that it becomes NS _ijt , and there are maximum I × J × T constraints.

3) Constraints for calculating the number of maintenance operations

The above-described constraint expression means that a minimum of 0 to a maximum of N actions can occur in each period t, and there are maximum T number of constraints.

The above constraint formula is a constraint formula that determines the number of jobs specifically generated in each period t, and there are a maximum of T constraint formulas.

4) Constraints for calculating the number of unavailable programs

At this time, Zit means whether or not an action has been taken for train i organization in period t (check 1, otherwise 0).

The above-described constraint formula is a constraint formula for calculating formation i as an unavailable formation according to the action task when an action work occurs on a tire attached to formation i in each period t, and there are a maximum of I × T number of constraint expressions.

The above constraint expression means that a minimum of 0 to a maximum of I unavailable units can occur in each period t, and there are a maximum of T number of constraints.

The above constraint equation is a constraint equation for determining the number of unavailable teams in each period t, and there are a maximum of T constraint equations.

5) Constraints for calculating the residual state score at the time of action

At this time, Ri _ijt means the remaining condition score before the action for the railway vehicle i organization j tires in period t.

The above-described constraint equation is a constraint equation that records the current state score as the remaining state score if an action on the railroad vehicle i formation j tire occurs in each period t, and records 0 if no action occurs. Maximum I × J × T constraints exist.

Through these various constraints, it is possible to establish an optimized schedule plan for the maintenance of railway system parts according to various assumptions and purpose definitions.

The protection scope of the present invention is not limited to the description and expression of the embodiments explicitly described above. In addition, it is added once again that the protection scope of the present invention cannot be limited due to obvious changes or substitutions in the technical field to which the present invention belongs.

100: maintenance schedule calculator
110: data collection unit
120: state diagnosis prediction value calculation unit
130: maintenance schedule calculation unit

Claims (14)

In the maintenance schedule calculation method for calculating the maintenance schedule of the parts constituting the railway system,
collecting diagnostic score data on the condition of rail system components;
Calculating a condition diagnosis prediction value of a railway system component from the diagnosis score data;
Calculating a maintenance schedule for railway system parts from the condition diagnosis prediction value;
Calculating the predicted value,
Calculating a condition diagnosis prediction value through an ARIMA (AutoRegressive Integrated Moving Average) model from the diagnosis score data,
The step of establishing the ARIMA model,
confirming the normality of the score data - calculating normal data by difference in case of abnormal data;
Defining and initial setting a prediction period and minAIC (Akaike Information Criterion);
Deriving time series data according to an autoregressive order;
Estimating ARMA model parameters (p, q) through Maximum Likelihood Estimation (MLE) method;
Calculating AIC (p, q), which is the goodness of fit of the model - the AIC (p, q) is defined by the formula below, where

is the variance of the prediction error, and N is the number of data when establishing the model.

-;
If the AIC(p, q) is less than minAIC and the degrees p and q are greater than or equal to log(N), calculating a state diagnosis predicted value of the next step 1; and
Characterized in that it includes repeating the step of calculating the condition diagnosis prediction value from the step of arranging the forecast period, p, q by one step until the defined prediction period, and arranging them into the time series.
A method for calculating an optimized maintenance schedule for railway system components. delete delete According to claim 1,
Calculating the maintenance schedule,
Characterized in that the maintenance schedule is calculated by considering all of the minimization of unavailability of railway vehicles, the minimization of the residual condition diagnosis prediction value before maintenance, and the minimization of maintenance work occurring during the same period,
A method for calculating an optimized maintenance schedule for railway system components. According to claim 4,
Characterized in that the objective function for calculating the maintenance schedule is defined by the following equation,
A method for calculating an optimized maintenance schedule for railway system components.
MIN

At this time,
i: index for rolling stock formation, i=1, 2, … I
j: Index of tires for each train group, j = 1, 2, . . . J
k: Index of the number of unavailable trains, k=0, 1, 2, … I
(Maximum cost of ownership = total number of trains)
n: Index of the number of action tasks, n=0, 1, 2, … N
t : index for period, t=1, 2, … T
P ¹ _k : Penalty weight according to the number of unavailable formations
D _kt : variable that determines the number of unavailable trains in period t
(D _kt =1 if k flight unavailable, otherwise 0)
^P2 : Weight
R _ijt : Remaining condition score before action for railroad car i organization j tire in period t
P ³ _n: Weight
Y _nt : means a variable that determines the number of action tasks in period t (Y _nt =1 for n tasks, otherwise 0). According to claim 1,
The diagnostic score data,
It is characterized by collecting diagnosis score data on the condition of railway system parts, which means the condition diagnosis score for railway system parts obtained through actual measurement,
A method for calculating an optimized maintenance schedule for railway system components. According to claim 1,
The railway system parts,
Characterized in that it is a light rail tire,
A method for calculating an optimized maintenance schedule for railway system components. a data collection unit that collects diagnostic score data for parts states in order to predict the state of railway system parts;
a condition diagnosis prediction value calculation unit that predicts condition diagnosis of railway system components for a certain period from the diagnosis score data; and
A maintenance schedule calculation unit that calculates a maintenance schedule for railway system parts through an objective function from the condition diagnosis prediction value,
The state diagnosis prediction value calculation unit,
Calculating a condition diagnosis prediction value through an ARIMA (AutoRegressive Integrated Moving Average) model from the diagnosis score data,
The method of establishing the ARIMA model,
confirming the normality of the score data - calculating normal data by difference in case of abnormal data;
Defining and initial setting a prediction period and minAIC (Akaike Information Criterion);
Deriving time series data according to an autoregressive order;
Estimating ARMA model parameters (p, q) through Maximum Likelihood Estimation (MLE) method;
Calculating AIC (p, q), which is the goodness of fit of the model - the AIC (p, q) is defined by the formula below, where

is the variance of the prediction error, and N is the number of data when establishing the model.

-;
If the AIC(p, q) is less than minAIC and the degrees p and q are greater than or equal to log(N), calculating a state diagnosis predicted value of the next step 1; and
Characterized in that it includes repeating the step of calculating the condition diagnosis prediction value from the step of arranging the forecast period, p, q by one step until the defined prediction period, and arranging them into the time series.
Optimized maintenance scheduling calculator for rail system components. delete delete According to claim 8,
The maintenance schedule calculator,
Characterized in that the maintenance schedule is calculated by considering all of the minimization of unavailability of railway vehicles, the minimization of the residual condition diagnosis prediction value before maintenance, and the minimization of maintenance work occurring during the same period,
Optimized maintenance scheduling calculator for rail system components. According to claim 11,
Characterized in that the objective function for calculating the maintenance schedule is defined by the following equation,
Optimized maintenance scheduling calculator for rail system components.
MIN

At this time,
i: index for rolling stock formation, i=1, 2, … I
j: Index of tires for each train group, j = 1, 2, . . . J
k: Index of the number of unavailable trains, k=0, 1, 2, … I
(Maximum cost of ownership = total number of trains)
n: Index of the number of action tasks, n=0, 1, 2, … N
t : index for period, t=1, 2, … T
P ¹ _k : Penalty weight according to the number of unavailable formations
D _kt : variable that determines the number of unavailable trains in period t
(D _kt =1 if k flight unavailable, otherwise 0)
^P2 : Weight
R _ijt : Remaining condition score before action for railroad car i organization j tire in period t
P ³ _n: Weight
Y _nt : means a variable that determines the number of action tasks in period t (Y _nt =1 for n tasks, otherwise 0). According to claim 8,
The diagnostic score data,
It is characterized by collecting diagnosis score data on the condition of railway system parts, which means the condition diagnosis score for railway system parts obtained through actual measurement,
Optimized maintenance scheduling calculator for rail system components. According to claim 8,
The railway system parts,
Characterized in that it is a light rail tire,
Optimized maintenance scheduling calculator for rail system components.

Images