Suggested Add-Ons – Zone Analysis

Introduction

The following is an extension of a previous itemset analysis done last month for Marine Fasteners customers over the course of the company’s history. Here, we are looking at data from just the previous five years and according to each customer Zone, or industry. As with the previous study, this analysis looks at item families rather than specific SKUs, as the particular dimensions of parts will vary wildly between customers according to their needs.

The data for this analysis was mined and cleaned in the same way as the previous study. I have shown the code for the first zone, but hidden it for all the others for cleanliness.

Zone by zone, we’ll look into what the algorithm can show us…

Zone B - Brikksen

b_basket = read.transactions("W:/zone analysis/baskets/b_basket.csv", sep = ',', rm.duplicates = TRUE)
summary(b_basket)

## transactions as itemMatrix in sparse format with
##  1096 rows (elements/itemsets/transactions) and
##  444 columns (items) and a density of 0.05301424 
## 
## most frequent items:
##   HHCSS   SHCSS   PPMSS   BSCSS   NFISS (Other) 
##     761     618     547     524     520   22828 
## 
## element (itemset/transaction) length distribution:
## sizes
##   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17  18  19  20 
## 138  86  56  63  34  29  35  31  26  25  28  18  15  15  21  15  16  14  17   9 
##  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40 
##   8  15   8  14  14  13  12   7   8  10   6   8   6   8   9   6   9   2   3   7 
##  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60 
##   7   6   8   4   3   9   2   2   6   7   7   7   5   4   8   7   2   2   5   2 
##  61  62  63  64  65  66  67  68  69  70  71  73  74  75  76  77  78  80  81  82 
##   4   6   4   3   4   2   4   5   1   6   7   4   2   1   3   3   1   1   2   2 
##  83  84  85  86  87  89  90  91  92  93  95  96  99 100 101 103 105 109 113 119 
##   2   5   1   3   4   3   3   2   3   3   2   1   1   1   5   2   2   2   2   2 
## 120 121 122 124 126 127 128 130 135 136 140 143 158 
##   1   1   1   1   1   2   2   1   1   1   1   1   1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00    3.00   11.00   23.54   34.00  158.00 
## 
## includes extended item information - examples:
##   labels
## 1  11205
## 2   1301
## 3   1401

itemFrequencyPlot(b_basket, topN = 20, main = "Zone B", xlab = "most frequent items", ylab = "")

b_rules = eclat(data = b_basket, parameter = list(support = 0.2, minlen = 2, maxlen = 2))

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE     0.2      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 219 
## 
## create itemset ... 
## set transactions ...[444 item(s), 1096 transaction(s)] done [0.00s].
## sorting and recoding items ... [35 item(s)] done [0.00s].
## creating bit matrix ... [35 row(s), 1096 column(s)] done [0.00s].
## writing  ... [336 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

inspect(sort(b_rules, by = 'support')[1:25])

##      items          support   count
## [1]  {HHCSS, SHCSS} 0.4708029 516  
## [2]  {HHCSS, NFISS} 0.4507299 494  
## [3]  {HHCSS, PPMSS} 0.4425182 485  
## [4]  {HHCSS, WFLSS} 0.4416058 484  
## [5]  {HHCSS, WMLSS} 0.4361314 478  
## [6]  {ESN79, HHCSS} 0.4352190 477  
## [7]  {BSCSS, HHCSS} 0.4160584 456  
## [8]  {PPMSS, SHCSS} 0.4069343 446  
## [9]  {BSCSS, SHCSS} 0.4051095 444  
## [10] {NFISS, WFLSS} 0.3959854 434  
## [11] {FSCSS, HHCSS} 0.3932482 431  
## [12] {HHCSS, PFMSS} 0.3841241 421  
## [13] {FSCSS, SHCSS} 0.3813869 418  
## [14] {NFISS, WMLSS} 0.3813869 418  
## [15] {WFLSS, WMLSS} 0.3759124 412  
## [16] {ESN79, NFISS} 0.3759124 412  
## [17] {SHCSS, WMLSS} 0.3722628 408  
## [18] {ESN79, SHCSS} 0.3722628 408  
## [19] {ESN79, WFLSS} 0.3713504 407  
## [20] {HHCSS, PPASS} 0.3704380 406  
## [21] {BSCSS, PPMSS} 0.3658759 401  
## [22] {ESN79, PPMSS} 0.3649635 400  
## [23] {ESN79, WMLSS} 0.3640511 399  
## [24] {NFISS, SHCSS} 0.3640511 399  
## [25] {PFMSS, PPMSS} 0.3631387 398

Zone B Summary

– 1,096 unique customers
– Average customer purchases ~23 unique parts over five years
– Top Five Items:

Hex Head Cap Screws
Socket Head Cap Screws
Phillips Pan Machine Screws
Button Socket Cap Screws
Hex Finish Nuts

– Moderate purchasing correlation between all parts in the Top Five (35-47%)

Zone CT - Cooling Towers

## transactions as itemMatrix in sparse format with
##  6 rows (elements/itemsets/transactions) and
##  48 columns (items) and a density of 0.3541667 
## 
## most frequent items:
##   HHCSS   HJTEK   SFTSS   WFLSS   WFLSW (Other) 
##       6       5       5       5       5      76 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1 14 16 20 23 28 
##  1  1  1  1  1  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00   14.50   18.00   17.00   22.25   28.00 
## 
## includes extended item information - examples:
##   labels
## 1   1401
## 2  ANCWD
## 3  BCGSS

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE     0.2      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 1 
## 
## create itemset ... 
## set transactions ...[48 item(s), 6 transaction(s)] done [0.00s].
## sorting and recoding items ... [22 item(s)] done [0.00s].
## creating bit matrix ... [22 row(s), 6 column(s)] done [0.00s].
## writing  ... [193 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {HHCSS, WFLSS} 0.8333333 5    
## [2]  {HJTEK, WFLSS} 0.8333333 5    
## [3]  {SFTSS, WFLSS} 0.8333333 5    
## [4]  {WFLSS, WFLSW} 0.8333333 5    
## [5]  {HHCSS, WFLSW} 0.8333333 5    
## [6]  {HJTEK, WFLSW} 0.8333333 5    
## [7]  {SFTSS, WFLSW} 0.8333333 5    
## [8]  {HHCSS, SFTSS} 0.8333333 5    
## [9]  {HJTEK, SFTSS} 0.8333333 5    
## [10] {HHCSS, HJTEK} 0.8333333 5    
## [11] {HHCSS, NFISS} 0.6666667 4    
## [12] {HJTEK, NFISS} 0.6666667 4    
## [13] {NFISS, SFTSS} 0.6666667 4    
## [14] {NFISS, WFLSW} 0.6666667 4    
## [15] {NFISS, WFLSS} 0.6666667 4    
## [16] {ESN09, HHCSS} 0.6666667 4    
## [17] {ESN09, HJTEK} 0.6666667 4    
## [18] {ESN09, SFTSS} 0.6666667 4    
## [19] {ESN09, WFLSW} 0.6666667 4    
## [20] {ESN09, WFLSS} 0.6666667 4    
## [21] {ESN09, WFESW} 0.6666667 4    
## [22] {HHCSS, WFESW} 0.6666667 4    
## [23] {HJTEK, WFESW} 0.6666667 4    
## [24] {SFTSS, WFESW} 0.6666667 4    
## [25] {WFESW, WFLSW} 0.6666667 4

Zone CT Summary

– Only 6 unique customers
– Average customer purchases ~17 unique parts over five years
– Top Five Items:

Hex Head Cap Screws
Hex Tek Washers
Stainless Full Threaded Studs
Stainless Flat Washers
316 Flat Washers

– Very strong purchasing correlation between all itemsets in the affinity algorithm (66-83%, highest of all zones)

Zone CW - Curtain Wall

## transactions as itemMatrix in sparse format with
##  16 rows (elements/itemsets/transactions) and
##  175 columns (items) and a density of 0.1353571 
## 
## most frequent items:
##   WFLSS   PFASS   PPASS   HHCSS   NFISS (Other) 
##      11      10       9       8       8     333 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1  2  3  4  5  7 16 17 24 31 33 42 51 62 76 
##  1  1  1  1  2  1  1  1  1  1  1  1  1  1  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00    4.75   16.50   23.69   35.25   76.00 
## 
## includes extended item information - examples:
##   labels
## 1   1401
## 2  ACCZ4
## 3  ACCZ5

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE     0.2      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 3 
## 
## create itemset ... 
## set transactions ...[175 item(s), 16 transaction(s)] done [0.00s].
## sorting and recoding items ... [28 item(s)] done [0.00s].
## creating bit matrix ... [28 row(s), 16 column(s)] done [0.00s].
## writing  ... [173 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support count
## [1]  {PPASS, WFLSS} 0.5625  9    
## [2]  {PFASS, PPASS} 0.5625  9    
## [3]  {PFASS, WFLSS} 0.5625  9    
## [4]  {NFISS, WFLSS} 0.5000  8    
## [5]  {PPMSS, WFLSS} 0.5000  8    
## [6]  {PFASS, PPMSS} 0.5000  8    
## [7]  {PPASS, PPMSS} 0.5000  8    
## [8]  {HHCSS, WFLSS} 0.5000  8    
## [9]  {HHCSS, PFASS} 0.5000  8    
## [10] {HHCSS, PPASS} 0.5000  8    
## [11] {WFLSS, WMLSS} 0.4375  7    
## [12] {PFMSS, WFLSS} 0.4375  7    
## [13] {PFASS, PFMSS} 0.4375  7    
## [14] {PFMSS, PPASS} 0.4375  7    
## [15] {PFMSS, PPMSS} 0.4375  7    
## [16] {ESN79, WFLSS} 0.4375  7    
## [17] {ESN79, PFASS} 0.4375  7    
## [18] {ESN79, PPASS} 0.4375  7    
## [19] {ESN79, HHCSS} 0.4375  7    
## [20] {NFISS, PFASS} 0.4375  7    
## [21] {NFISS, PPASS} 0.4375  7    
## [22] {HHCSS, NFISS} 0.4375  7    
## [23] {HJTEK, WFLSS} 0.4375  7    
## [24] {HJTEK, PFASS} 0.4375  7    
## [25] {HJTEK, PPASS} 0.4375  7

Zone CW Summary

– 16 unique customers
– Average customer purchases ~24 unique parts over five years
– Top Five Items:

Stainless Flat Washers
Phillips Flat-Head Screws
Phillips Pan-Head Screws
Hex Head Cap Screws
Hex Finish Nuts

– Moderate purchasing correlation between top 25 itemsets per the algorithm (44-56%)

Zone FENCE - Fencing

## transactions as itemMatrix in sparse format with
##  36 rows (elements/itemsets/transactions) and
##  113 columns (items) and a density of 0.05948869 
## 
## most frequent items:
##   HJTEK   PPASS   HOOKL   PPTEK   HHCSS (Other) 
##      23      15      14       9       7     174 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1  2  3  4  5  6  9 11 13 15 24 25 33 37 
##  8 10  2  4  2  1  2  1  1  1  1  1  1  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   1.000   2.000   2.500   6.722   6.750  37.000 
## 
## includes extended item information - examples:
##   labels
## 1   1101
## 2   1401
## 3  AIRFR

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 9 
## 
## create itemset ... 
## set transactions ...[113 item(s), 36 transaction(s)] done [0.00s].
## sorting and recoding items ... [4 item(s)] done [0.00s].
## creating bit matrix ... [4 row(s), 36 column(s)] done [0.00s].
## writing  ... [1 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##     items          support count
## [1] {HJTEK, PPASS} 0.25    9

Zone FENCE Summary

– 36 unique customers
– Average customer purchases ~7 unique parts over five years
– Very little purchasing patterns
– Algorithm only found one itemset (HJTEK, PPASS) which were purchased together 9 times

Zone H20 - Pumps, Valves, Etc.

## transactions as itemMatrix in sparse format with
##  3 rows (elements/itemsets/transactions) and
##  12 columns (items) and a density of 0.4444444 
## 
## most frequent items:
##   HHCSS   MS157   NFISS   WFLSS    1401 (Other) 
##       2       2       2       2       1       7 
## 
## element (itemset/transaction) length distribution:
## sizes
## 1 6 9 
## 1 1 1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   1.000   3.500   6.000   5.333   7.500   9.000 
## 
## includes extended item information - examples:
##   labels
## 1   1401
## 2  BOLTS
## 3  ESN79

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 0 
## 
## create itemset ... 
## set transactions ...[12 item(s), 3 transaction(s)] done [0.00s].
## sorting and recoding items ... [12 item(s)] done [0.00s].
## creating bit matrix ... [12 row(s), 3 column(s)] done [0.00s].
## writing  ... [48 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {HHCSS, WFLSS} 0.6666667 2    
## [2]  {NFISS, WFLSS} 0.6666667 2    
## [3]  {HHCSS, NFISS} 0.6666667 2    
## [4]  {HHCSS, WFLSW} 0.3333333 1    
## [5]  {NFISS, WFLSW} 0.3333333 1    
## [6]  {WFLSS, WFLSW} 0.3333333 1    
## [7]  {BOLTS, WFLSW} 0.3333333 1    
## [8]  {NFISW, WFLSW} 0.3333333 1    
## [9]  {HHCSS, NFISW} 0.3333333 1    
## [10] {NFISS, NFISW} 0.3333333 1    
## [11] {NFISW, WFLSS} 0.3333333 1    
## [12] {BOLTS, NFISW} 0.3333333 1    
## [13] {BOLTS, HHCSS} 0.3333333 1    
## [14] {BOLTS, NFISS} 0.3333333 1    
## [15] {BOLTS, WFLSS} 0.3333333 1    
## [16] {HHCSS, WUSSZ} 0.3333333 1    
## [17] {NFISS, WUSSZ} 0.3333333 1    
## [18] {WFLSS, WUSSZ} 0.3333333 1    
## [19] {MS157, WUSSZ} 0.3333333 1    
## [20] {1401, WUSSZ}  0.3333333 1    
## [21] {ESN79, WUSSZ} 0.3333333 1    
## [22] {NTESS, WUSSZ} 0.3333333 1    
## [23] {WFESZ, WUSSZ} 0.3333333 1    
## [24] {HHCSS, WFESZ} 0.3333333 1    
## [25] {NFISS, WFESZ} 0.3333333 1

Zone H20 Summary

– Only 3 unique customers (relatively new zone)
– Not much data to analyze
– Very little purchasing patterns
– Due to small size of dataset, algorithm found purchasing correlations between almost all parts purchased

Zone I - Industrial

## transactions as itemMatrix in sparse format with
##  1018 rows (elements/itemsets/transactions) and
##  1115 columns (items) and a density of 0.00919679 
## 
## most frequent items:
##   HHCSS   WFLSS   NFISS   ESN79   PPMSS (Other) 
##     443     401     326     306     286    8677 
## 
## element (itemset/transaction) length distribution:
## sizes
##   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17  18  19  20 
## 205 114  89  59  56  56  36  36  33  16  12  17  21  18  15  15  12  17  12  11 
##  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  38  39  40  41 
##  13  17  15  13  11  12   7   6   4   6   3   2   3   4   5   3   3   1   4   5 
##  42  43  44  45  46  47  49  50  51  52  54  55  56  57  60  62  76  86  88  94 
##   1   1   1   1   2   2   2   2   1   2   1   1   1   3   2   1   1   1   3   1 
## 118 
##   1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00    2.00    5.00   10.25   14.00  118.00 
## 
## includes extended item information - examples:
##   labels
## 1  1/2BU
## 2  10000
## 3  10001

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 254 
## 
## create itemset ... 
## set transactions ...[1115 item(s), 1018 transaction(s)] done [0.00s].
## sorting and recoding items ... [5 item(s)] done [0.00s].
## creating bit matrix ... [5 row(s), 1018 column(s)] done [0.00s].
## writing  ... [3 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##     items          support   count
## [1] {HHCSS, WFLSS} 0.3133595 319  
## [2] {HHCSS, NFISS} 0.2760314 281  
## [3] {NFISS, WFLSS} 0.2760314 281

Zone I Summary

– 1,018 unique customers
– Average customer purchases ~10 unique parts over five years
– Top Five Items:

Hex Head Cap Screws
Stailness Flat Washers
Hex Finish Nuts
Nylon Insert Lock Nuts
Phillips Pan Machine Screws

– Algorithm only found three relatively weak purchasing correlations

Zone M - Marine

## transactions as itemMatrix in sparse format with
##  714 rows (elements/itemsets/transactions) and
##  1103 columns (items) and a density of 0.02306671 
## 
## most frequent items:
##   HHCSS   ESN79   PPMSS   PPASS   PFMSS (Other) 
##     461     413     405     376     371   16140 
## 
## element (itemset/transaction) length distribution:
## sizes
##   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15  16  17  18  19  20 
##  81  43  39  27  20  24  19  20  13  16  16  18  11  16   9  11   8  10  11  16 
##  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40 
##   8  19  11   9   9   7   5   7   5   7   7   2  11   6   5  10   3   4   8   3 
##  41  42  43  44  45  46  47  48  49  50  51  53  54  55  56  57  58  59  60  61 
##   2   6   3   4   4   2   6   6   5   3   2   2   1   2   2   2   2   3   1   1 
##  62  63  66  67  68  69  71  72  74  75  79  80  82  83  84  85  86  87  88  89 
##   4   3   4   1   1   2   3   5   1   1   5   1   1   2   2   2   3   1   2   3 
##  90  91  94  99 100 101 103 105 106 108 109 111 113 119 121 122 134 137 145 148 
##   2   1   1   1   2   4   1   1   1   1   1   1   1   1   2   1   1   2   1   1 
## 154 159 163 181 215 233 274 
##   1   1   1   1   1   1   1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00    4.00   14.00   25.44   33.75  274.00 
## 
## includes extended item information - examples:
##   labels
## 1   1101
## 2  12655
## 3    129

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 178 
## 
## create itemset ... 
## set transactions ...[1103 item(s), 714 transaction(s)] done [0.00s].
## sorting and recoding items ... [21 item(s)] done [0.00s].
## creating bit matrix ... [21 row(s), 714 column(s)] done [0.00s].
## writing  ... [89 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {ESN79, HHCSS} 0.5070028 362  
## [2]  {HHCSS, PPMSS} 0.4915966 351  
## [3]  {ESN79, PPMSS} 0.4719888 337  
## [4]  {HHCSS, WFLSS} 0.4551821 325  
## [5]  {ESN79, WFLSS} 0.4523810 323  
## [6]  {PFMSS, PPMSS} 0.4523810 323  
## [7]  {PPASS, PPMSS} 0.4495798 321  
## [8]  {HHCSS, PFMSS} 0.4481793 320  
## [9]  {ESN79, PPASS} 0.4439776 317  
## [10] {HHCSS, PPASS} 0.4425770 316  
## [11] {ESN79, PFMSS} 0.4229692 302  
## [12] {PFASS, PPASS} 0.4173669 298  
## [13] {PFMSS, PPASS} 0.4159664 297  
## [14] {PPMSS, WFLSS} 0.4145658 296  
## [15] {PFASS, PPMSS} 0.4019608 287  
## [16] {HHCSS, PFASS} 0.3977591 284  
## [17] {PPASS, WFLSS} 0.3963585 283  
## [18] {PFASS, PFMSS} 0.3921569 280  
## [19] {ESN79, PFASS} 0.3879552 277  
## [20] {POASS, PPASS} 0.3767507 269  
## [21] {POMSS, PPMSS} 0.3767507 269  
## [22] {HHCSS, NFISS} 0.3753501 268  
## [23] {HHCSS, POMSS} 0.3697479 264  
## [24] {ESN79, NFISS} 0.3655462 261  
## [25] {PFMSS, WFLSS} 0.3655462 261

Zone M Summary

– 714 unique customers
– Average customer purchases ~25 unique parts over five years
– Top Five Items:

Hex Head Cap Screws
Nylon Insert Lock Nuts
Phillips Pan Machine Screws
Phillips Pan Type A Screws
Phillips Flat Machine Screws

– Algorithm only found moderate purchasing correlations between almost all parts in the top 20 (35-50%)

Zone MD - Marine Distributors

## transactions as itemMatrix in sparse format with
##  30 rows (elements/itemsets/transactions) and
##  160 columns (items) and a density of 0.13625 
## 
## most frequent items:
##   HHCSS   NFISS   BHLSS   ESN79   PFASS (Other) 
##      23      22      21      21      21     546 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1  3  4  5  6  7  8  9 13 15 16 18 19 20 23 25 27 28 29 30 31 32 38 40 45 47 
##  1  2  1  1  1  1  1  1  1  1  1  1  1  1  1  2  1  1  1  2  1  1  1  1  1  1 
## 57 
##  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00    8.25   21.50   21.80   30.00   57.00 
## 
## includes extended item information - examples:
##   labels
## 1  11205
## 2   1401
## 3    300

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 7 
## 
## create itemset ... 
## set transactions ...[160 item(s), 30 transaction(s)] done [0.00s].
## sorting and recoding items ... [24 item(s)] done [0.00s].
## creating bit matrix ... [24 row(s), 30 column(s)] done [0.00s].
## writing  ... [219 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {ESN79, NFISS} 0.6666667 20   
## [2]  {HHCSS, NFISS} 0.6666667 20   
## [3]  {ESN79, HHCSS} 0.6333333 19   
## [4]  {NFISS, PFASS} 0.6333333 19   
## [5]  {PFASS, PPASS} 0.6333333 19   
## [6]  {HHCSS, PPASS} 0.6333333 19   
## [7]  {NFISS, PPASS} 0.6333333 19   
## [8]  {BHLSS, HHCSS} 0.6333333 19   
## [9]  {POASS, PPASS} 0.6000000 18   
## [10] {ESN79, PPASS} 0.6000000 18   
## [11] {HHCSS, PFASS} 0.6000000 18   
## [12] {BHLSS, NFISS} 0.6000000 18   
## [13] {NFISS, PPMSS} 0.5666667 17   
## [14] {BHLSS, WFISW} 0.5666667 17   
## [15] {HHCSS, POASS} 0.5666667 17   
## [16] {NFISS, POASS} 0.5666667 17   
## [17] {HHCSS, WFLSS} 0.5666667 17   
## [18] {NFISS, WFLSS} 0.5666667 17   
## [19] {ESN79, WFLSS} 0.5666667 17   
## [20] {BHLSS, ESN79} 0.5666667 17   
## [21] {ESN79, PFASS} 0.5666667 17   
## [22] {BHLSS, PFASS} 0.5666667 17   
## [23] {BHLSS, PPASS} 0.5666667 17   
## [24] {PFASS, PPMSS} 0.5333333 16   
## [25] {HHCSS, WFISW} 0.5333333 16

Zone MD Summary

– 30 unique customers
– Average customer purchases ~22 unique parts over five years
– Top Five Items:

Hex Head Cap Screws
Hex Finish Nuts
Hex Lag Bolts
Nylon Insert Lock Nuts
Phillips Flat Type A Screws

– Algorithm found strong purchasing correlations (53-66%)

Zone ODF - Outdoor Furniture

## transactions as itemMatrix in sparse format with
##  68 rows (elements/itemsets/transactions) and
##  170 columns (items) and a density of 0.06980969 
## 
## most frequent items:
##   QFDE1   ESN79   WFLSS   HHCSS   BHLSS (Other) 
##      40      38      37      33      30     629 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 17 18 19 20 22 23 24 26 28 29 36 
##  2  4  4  4  4  4  3  2  1  4  2  5  4  4  3  2  2  2  4  1  1  1  1  2  1  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    1.00    5.00   11.50   11.87   17.00   36.00 
## 
## includes extended item information - examples:
##   labels
## 1   1301
## 2   1401
## 3  32085

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 17 
## 
## create itemset ... 
## set transactions ...[170 item(s), 68 transaction(s)] done [0.00s].
## sorting and recoding items ... [11 item(s)] done [0.00s].
## creating bit matrix ... [11 row(s), 68 column(s)] done [0.00s].
## writing  ... [22 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {ESN79, WFLSS} 0.4705882 32   
## [2]  {ESN79, HHCSS} 0.3823529 26   
## [3]  {QFDE1, WFLSS} 0.3823529 26   
## [4]  {BCGSS, WFLSS} 0.3676471 25   
## [5]  {BHLSS, WFLSS} 0.3529412 24   
## [6]  {BHLSS, ESN79} 0.3529412 24   
## [7]  {HHCSS, WFLSS} 0.3235294 22   
## [8]  {ESN79, QFDE1} 0.3235294 22   
## [9]  {QFBG1, QFDE1} 0.3088235 21   
## [10] {BCGSS, ESN79} 0.3088235 21   
## [11] {NFISS, WFLSS} 0.2941176 20   
## [12] {BCGSS, QFDE1} 0.2941176 20   
## [13] {BHLSS, QFDE1} 0.2941176 20   
## [14] {QFBG1, WFLSS} 0.2794118 19   
## [15] {HHCSS, QFDE1} 0.2794118 19   
## [16] {ESN79, NFISS} 0.2647059 18   
## [17] {ESN79, QFBG1} 0.2647059 18   
## [18] {BCGSS, BHLSS} 0.2647059 18   
## [19] {BHLSS, HHCSS} 0.2647059 18   
## [20] {QBDE1, WFLSS} 0.2500000 17   
## [21] {ESN79, QBDE1} 0.2500000 17   
## [22] {QBDE1, QFDE1} 0.2500000 17

Zone ODF Summary

– 68 unique customers
– Average customer purchases ~12 unique parts over five years
– Top Five Items:

Square Flat Deck Screws Type 17 (only Top Five appearance)
Nylon Insert Lock Nuts
Stainless Flat Washers
Hex Head Cap Screws
Hex Lag Bolts

– Algorithm found moderate purchasing correlations (25-47%)

Zone S - Solar

## transactions as itemMatrix in sparse format with
##  107 rows (elements/itemsets/transactions) and
##  270 columns (items) and a density of 0.03243337 
## 
## most frequent items:
##   HHCSS   WFLSS   NFLSS   BFLSS   HJTEK (Other) 
##      57      46      41      33      33     727 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1  2  3  4  5  6  7  8  9 11 12 13 14 15 16 17 19 21 22 23 24 25 26 28 33 36 
## 12  9 12 14  8 13  2  2  4  1  4  2  2  3  2  2  2  2  1  1  1  1  1  2  1  1 
## 37 45 
##  1  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   1.000   3.000   5.000   8.757  12.000  45.000 
## 
## includes extended item information - examples:
##   labels
## 1  11209
## 2   1401
## 3  21401

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 26 
## 
## create itemset ... 
## set transactions ...[270 item(s), 107 transaction(s)] done [0.00s].
## sorting and recoding items ... [9 item(s)] done [0.00s].
## creating bit matrix ... [9 row(s), 107 column(s)] done [0.00s].
## writing  ... [5 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##     items          support   count
## [1] {HHCSS, WFLSS} 0.3738318 40   
## [2] {HHCSS, NFLSS} 0.2990654 32   
## [3] {HHCSS, WMLSS} 0.2803738 30   
## [4] {NFLSS, WFLSS} 0.2616822 28   
## [5] {WFLSS, WMLSS} 0.2523364 27

Zone S Summary

– 107 unique customers
– Average customer purchases ~9 unique parts over five years
– Top Five Items:

Hex Head Cap Screws
Stainless Flat Washers
Hex Flange Nuts (only Top Five appearance)
Hex Flange Bolts (only Top Five appearance)
Hex Tek Washers

– Algorithm found very few, relatively weak purchasing correlations (25-38%)

Zone T&L - Trailers and Lifts

## transactions as itemMatrix in sparse format with
##  35 rows (elements/itemsets/transactions) and
##  158 columns (items) and a density of 0.0835443 
## 
## most frequent items:
##   ESN79   HHCSS   WFLSS   BCGSS   BHLSS (Other) 
##      25      22      22      16      16     361 
## 
## element (itemset/transaction) length distribution:
## sizes
##  1  2  3  4  5  6  7  8  9 10 12 13 14 15 22 29 30 35 51 
##  3  4  1  2  1  3  1  1  1  1  2  3  2  2  1  1  3  2  1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##     1.0     4.0    10.0    13.2    15.0    51.0 
## 
## includes extended item information - examples:
##   labels
## 1   1401
## 2  3-Mar
## 3  32085

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 8 
## 
## create itemset ... 
## set transactions ...[158 item(s), 35 transaction(s)] done [0.00s].
## sorting and recoding items ... [11 item(s)] done [0.00s].
## creating bit matrix ... [11 row(s), 35 column(s)] done [0.00s].
## writing  ... [26 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {ESN79, HHCSS} 0.6285714 22   
## [2]  {ESN79, WFLSS} 0.6000000 21   
## [3]  {HHCSS, WFLSS} 0.5714286 20   
## [4]  {BCGSS, ESN79} 0.4285714 15   
## [5]  {ESN79, NFISS} 0.4285714 15   
## [6]  {NFISS, WFLSS} 0.4285714 15   
## [7]  {BHLSS, ESN79} 0.4000000 14   
## [8]  {BHLSS, WFLSS} 0.4000000 14   
## [9]  {HHCSS, NFISS} 0.4000000 14   
## [10] {BCGSS, HHCSS} 0.3714286 13   
## [11] {BCGSS, WFLSS} 0.3714286 13   
## [12] {BHLSS, HHCSS} 0.3714286 13   
## [13] {ESN79, WMLSS} 0.3142857 11   
## [14] {HHCSS, WMLSS} 0.3142857 11   
## [15] {WFLSS, WMLSS} 0.3142857 11   
## [16] {NFISS, WMLSS} 0.3142857 11   
## [17] {BCGSS, NFISS} 0.3142857 11   
## [18] {BUQSS, ESN79} 0.2857143 10   
## [19] {BUQSS, WFLSS} 0.2857143 10   
## [20] {BUQSS, LUGSS} 0.2571429  9   
## [21] {ESN79, RIVAD} 0.2571429  9   
## [22] {ESN79, PPMSS} 0.2571429  9   
## [23] {PPMSS, WFLSS} 0.2571429  9   
## [24] {BUQSS, HHCSS} 0.2571429  9   
## [25] {BCGSS, BHLSS} 0.2571429  9   
## [26] {BHLSS, NFISS} 0.2571429  9

Zone T&L Summary

– 35 unique customers
– Average customer purchases ~13 unique parts over five years
– Top Five Items:

Nylon Insert Lock Nuts
Hex Head Cap Screws
Stainless Flat Washers
Stainless Carriage Bolts (only Top Five appearance)
Hex Lag Bolts

– Algorithm found moderate purchasing correlations (25-63%)

Zone W - Wurth Companies

## transactions as itemMatrix in sparse format with
##  18 rows (elements/itemsets/transactions) and
##  264 columns (items) and a density of 0.2272727 
## 
## most frequent items:
##   NFISS   HHCSS   WFLSS   MS157   NFISW (Other) 
##      17      15      15      14      14    1005 
## 
## element (itemset/transaction) length distribution:
## sizes
##   2   7   8  12  18  21  23  31  45  46  49  91 118 123 131 152 180 
##   1   1   1   1   1   1   2   1   1   1   1   1   1   1   1   1   1 
## 
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    2.00   18.75   38.00   60.00  111.25  180.00 
## 
## includes extended item information - examples:
##   labels
## 1   1301
## 2   1401
## 3  2PPAP

## Eclat
## 
## parameter specification:
##  tidLists support minlen maxlen            target  ext
##     FALSE    0.25      2      2 frequent itemsets TRUE
## 
## algorithmic control:
##  sparse sort verbose
##       7   -2    TRUE
## 
## Absolute minimum support count: 4 
## 
## create itemset ... 
## set transactions ...[264 item(s), 18 transaction(s)] done [0.00s].
## sorting and recoding items ... [95 item(s)] done [0.00s].
## creating bit matrix ... [95 row(s), 18 column(s)] done [0.00s].
## writing  ... [3791 set(s)] done [0.00s].
## Creating S4 object  ... done [0.00s].

##      items          support   count
## [1]  {NFISS, WFLSS} 0.8333333 15   
## [2]  {HHCSS, NFISS} 0.8333333 15   
## [3]  {NFISS, WMLSS} 0.7777778 14   
## [4]  {WFLSS, WMLSS} 0.7777778 14   
## [5]  {NFISS, WFLSW} 0.7777778 14   
## [6]  {HHCSS, WFLSW} 0.7777778 14   
## [7]  {NFISS, NFISW} 0.7777778 14   
## [8]  {MS157, NFISS} 0.7777778 14   
## [9]  {HHCSS, MS157} 0.7777778 14   
## [10] {NFISS, WFESS} 0.7222222 13   
## [11] {HHCSS, WMLSS} 0.7222222 13   
## [12] {HHCSS, WFLSS} 0.7222222 13   
## [13] {ESN09, NFISS} 0.7222222 13   
## [14] {ESN09, HHCSS} 0.7222222 13   
## [15] {ESN09, WFLSW} 0.7222222 13   
## [16] {MS157, WFLSW} 0.7222222 13   
## [17] {NFISW, WFLSW} 0.7222222 13   
## [18] {NFISS, PPMSS} 0.7222222 13   
## [19] {PPMSS, SHCSS} 0.7222222 13   
## [20] {NFISS, SHCSS} 0.7222222 13   
## [21] {HHCSS, NFISW} 0.7222222 13   
## [22] {MS157, NFISW} 0.7222222 13   
## [23] {FSCSS, NFISS} 0.6666667 12   
## [24] {FSCSS, WFLSS} 0.6666667 12   
## [25] {BSCSS, NFISS} 0.6666667 12

Zone W Summary

– 18 unique customers
– Average customer purchases ~60 unique parts over five years (highest of all zones by far)
– Top Five Items:

Hex Finish Nuts
Hex Head Cap Screws
Stainless Flat Washers
Non Standard Flat Washers
Hex Finish 316 Nuts

– Algorithm found very strong purchasing correlations (66-83%)

Conclusions / Next Steps

Unsurprisingly, there is a high prevalence of standard popular parts across almost all zones, consisting primarily of Hex Head Cap Screws, Phillips Pan/Flat Screws, Stainless Flat Washers, and Nylon Insert Lock Nuts. Of note is when this pattern deviates, particularly in Zone ODF, which features Square Flat Deck Screws and Hex Lag Bolts, and Zone S, which prominently shows Hex Flange Nuts and Bolts. Such deviations can greatly assist the sales team in providing specific part suggestions according to the customer’s industry.

As for the itemset findings, they largely align themselves with the Top 20 lists, as pairings among them abound throughout almost all zones. Some zones, such as FENCE, H20, I, and S, due to the way the data bears out, do not produce many part associations per the algorithm, but as time goes on, more data will accrue, and more correlations will be found in the future.

The next step for the PIDA team is to produce a tool that the sales team can use that will easily provide part suggestions according to the customer/zone with which they are working. Using the data from this analysis, as well as the initial Affinity Analysis and the Five Year-Over-Year Analysis accompanying this study, we are confident that this tool will prove to be a useful and profitable asset to the sales team moving forward.