智能网关>> Modified-Rank-Order-Clustering>> 返回
项目作者： nagdevAmruthnath

项目描述 ：
Includes a package for rank order clustering and modified rank order clustering proposed by Amruthnath (2016)
高级语言： R
项目地址: git://github.com/nagdevAmruthnath/Modified-Rank-Order-Clustering.git


Binder

Modified Rank Order Clustering

This REPO contains source code (in -mod—Rank-order-clustering/ROC/R/ROC.R) as well as package file (ROC_0.1.0.zip file in root) for Modified & Rank Order clustering. This package contains two functions ROC() and MROC(). ROC function runs based on algorithm proposed by King in 1980. MROC function is based on the algorithm proposed by Amruthnath in 2016. The package includes documentation.

Installation of package

The following package has been tested by installing them directly through R

  1. # install devtools
  2. install.packages("devtools")
  4. # load library
  5. library(devtools)
  7. # install MROC package
  8. devtools::install_github("nagdevAmruthnath/Modified-Rank-Order-Clustering")
  10. # Load ROC library
  11. library(ROC)

Create Machine Component Data

  1. # A function to create a random data matrix
  2. # Reference: http://dirk.eddelbuettel.com/blog/2012/09/02/#faster_binomial_matrix_creation
  4. createData = function(N, K) {
  5.   matrix(rbinom(N * K, 1, 0.5), ncol = K, nrow = N)
  6. }
  8. # use above function to create 10 x 10 matrix
  9. data = as.data.frame(createData(10,10))
  11. # view that random data
  12. data
  14. #    V1 V2 V3 V4 V5 V6 V7 V8 V9 V10
  15. # 1   0  1  0  0  0  1  1  0  0   1
  16. # 2   0  0  0  0  1  0  0  0  1   0
  17. # 3   0  0  1  0  0  1  0  1  0   1
  18. # 4   1  0  1  1  0  0  0  0  0   1
  19. # 5   0  1  1  0  1  0  1  1  0   1
  20. # 6   1  0  1  1  1  0  1  1  0   1
  21. # 7   1  1  0  0  1  1  0  0  0   0
  22. # 8   0  0  1  0  1  0  1  0  0   0
  23. # 9   1  1  1  0  1  0  1  1  0   1
  24. # 10  0  1  1  0  0  1  1  1  1   0

ROC: Rank Order Clustering

Description

This performs Rank Order Clustering matrix for production flow analysis. The result matrix can be used to cluster machines and part numbers

Usage

ROC(x)

Arguments

x is a data frame without column names or row names. Columns are partnumbers and rows are machines

Details

This function takes any number of machines or partnumbers to generate a Rank Order Clustering Matrix. This algorithm is based on Kings (1981) algorithm for production flow analysis. Maximum iterations are set to 50.

Value

result is a data frame consisting of final matrix from the algorithm

Note:

This algorithm is tested out for upto 100 x 100 matrix

Examples

  1. # load ROC library
  2. library(ROC)
  4. # run ROC algorithm
  5. ROC(data)
  6. #      1   5   2   3  10   7   8   6   4  9  V11
  7. # 9    1   1   1   1   1   1   1   0   0  0 1016
  8. # 7    1   1   1   0   0   0   0   1   0  0  900
  9. # 6    1   1   0   1   1   1   1   0   1  0  890
  10. # 4    1   0   0   1   1   0   0   0   1  0  610
  11. # 5    0   1   1   1   1   1   1   0   0  0  504
  12. # 8    0   1   0   1   0   1   0   0   0  0  336
  13. # 2    0   1   0   0   0   0   0   0   0  1  257
  14. # 10   0   0   1   1   0   1   1   1   0  1  221
  15. # 1    0   0   1   0   1   1   0   1   0  0  180
  16. # 3    0   0   0   1   1   0   1   1   0  0  108
  17. # 11 960 952 806 757 739 694 677 263 192 12   NA

MROC Modified Rank Order Clustering

Description

This function provides the Modified Rank Order Clustering result matrix. This can be used to cluster the machines and components.

Usage

MROC(data, compWeight, machineWeight)

Arguments

data: a data frame without row names and column names. Just include the machine component matrix
compWeight: a single data frame containing numerical component weighs
machineWeight: a single data frame containing numerical machine weighs

Details

This function is built for product flow in cellular manufacturing. This function is built based on Modified Rank Order Clustering proposed by Amruthnath and Gupta (2016). This algorithm is an extension of Rank Order Clustering but, it uses weight to reorder the initial matrix first.

Value
result: MROC matrix as a data frame

Examples

  1. # create random 10 weights for components
  2. compWeight = sample(1:100, 10)
  3. compWeight
  4. # [1] 100  74  50  85   7  46  87  92  36  59
  6. # create random 10 weights for machines
  7. machineWeight = sample(1:100, 10)
  8. machineWeight
  9. # [1] 98 37 48 50 15 23 18 19 77  2
  11. # run MROC algorithm with weights
  12. MROC(data, compWeight , machineWeight)
  13. #      X5  X9  X2  X1  X6  X3  X7  X8 X10 X4 V11
  14. # 2     1   1   0   0   0   0   0   0   0  0 768
  15. # 7     1   0   1   1   1   0   0   0   0  0 657
  16. # 9     1   0   1   1   0   1   1   1   1  0 631
  17. # 5     1   0   1   0   0   1   1   1   1  0 630
  18. # 6     1   0   0   1   0   1   1   1   1  1 623
  19. # 8     1   0   0   0   0   1   1   0   0  0 580
  20. # 10    0   1   1   0   1   1   1   1   0  0 470
  21. # 3     0   0   0   0   1   1   0   1   1  0 226
  22. # 1     0   0   1   0   1   0   1   0   1  0 180
  23. # 4     0   0   0   1   0   1   0   0   1  1 105
  24. # 11 1008 520 458 417 270 253 250 236 231 33  NA

References

J. R. KING (1980), Machine-component grouping in production flow analysis: an approach using a rank order clustering algorithm International Journal of Production Research Vol. 18 , Iss. 2,1980

Amruthnath, N., & Gupta, T. (2016). Modified Rank Order Clustering Algorithm Approach by Including Manufacturing Data. IFAC-PapersOnLine, 49(5), 138-142.