You are on page 1of 9

Quality Management Homework Assignment 1

Quality Management

Homework Assignment 1

Submitted by :
Group 4
Anand Singh (r0777287)
Amarin Neuville
Pranav Bhaskar Bhoyar
Ravi Ranjan
Sudeep Nayak

0|Page
Quality Management Homework Assignment 1

 Kindly refer the Excel file named “QM homework 1 group 4” for detailed computation.

(1) The lots allotted to the group for evaluation are as below (refer sheet “Lots Assigned”):
7 20 26 34 43 62 82 165 183 198

We apply the following parameters and calculate total cost in each case:
Case 1 : Single sampling plan, with n = 50; c = 1.
Case 2 : Double sampling plan, with n1 = 30; c1 = 0; r1 = 2; n2 = 24; c2 = 1.
Case 3 : Continuous sampling plan, with h1 = -1.08; h2 = 1.12; s=0.055.

Cost of Inspection : K1 = 1 EUR


Cost of using a defective unit : K2 = 150 EUR

We are taking the following assumption for cost calculation:


 For lots rejected by the respective sampling plans : Only the cost of sampling process
(K1) is incurred. Since these lots are not processed further the cost of using defective
units (K2) during assembly will not be incurred for such lots.

 For lots accepted by the respective sampling plans : Cost of sampling process (K1) is
incurred for all the lots. Defective units which remain undetected during sampling will
incur cost K2. Other defective units that were detected during the sampling process
are discarded and not used in assembly.

Case 1 (Single Sampling):


The units to be sampled were chosen at random (50 units). The list was created using Random
number generator (function ‘Randbetween’ of excel was used) (refer sheet “Units being
Sampled”). It has been summarised below:

Units selected for Single sampling plan


2 48 76 108 140
4 49 79 109 153
6 50 88 113 156
19 51 90 116 157
27 54 93 118 171
30 63 96 119 183
36 67 97 125 184
38 68 101 129 186
43 70 102 130 190
47 74 107 132 191

1|Page
Quality Management Homework Assignment 1

The summary of Single sampling process is provided below (refer sheet “Single Sampling”):

Lot No. of Non- No. of Defective


Accept/Reject
Number defective units units

7 49 1 ACCEPT
20 47 3 REJECT
26 45 5 REJECT
34 48 2 REJECT
43 46 4 REJECT
62 50 0 ACCEPT
82 49 1 ACCEPT
165 45 5 REJECT
183 49 1 ACCEPT
198 47 3 REJECT

We reject a lot if number of defective units is found out to be greater than 1 (c).
According to the decision to accept or reject a lot, the cost has been calculated as below:

Cost of Cost of processing


Total Cost
Lot Number Accept/Reject Sampling (If lot is accepted)
(EUR)
(EUR) (EUR)
7 ACCEPT 50 150 200
20 REJECT 50 0 50
26 REJECT 50 0 50
34 REJECT 50 0 50
43 REJECT 50 0 50
62 ACCEPT 50 1500 1550
82 ACCEPT 50 750 800
165 REJECT 50 0 50
183 ACCEPT 50 450 500
198 REJECT 50 0 50

Case 2 (Double Sampling):


Step 1: For the first set of units to be sampled, 30 units (n1) were chosen at random. The list
was created using Random number generator (refer sheet “Units being Sampled”). It has been
summarised below:

Units to be selected for first step of sampling


5 26 56 101 121 136
6 39 61 103 126 144
8 41 72 109 129 187
14 45 90 118 132 189

2|Page
Quality Management Homework Assignment 1

17 47 96 120 135 200


The summary of first step in Double sampling process is provided below (refer sheet “Double
Sampling”):

Lot No. of Non- No. of Defectives Step 1 result


number defectives (D1) (Accept|Reject|Undecided)

7 29 1 Undecided
20 28 2 Reject
26 30 0 Accept
34 27 3 Reject
43 28 2 Reject
62 28 2 Reject
82 30 0 Accept
165 26 4 Reject
183 29 1 Undecided
198 28 2 Reject

We accept a lot if number of defective units is found out to be lesser than or equal to 0 (c 1)
and reject the lot if number of defective units is found out to be greater than or equal to 2
(r1). For any other count of defective units we move to step 2 of double sampling for the lot.

Step 2: For the second set of units to be sampled, 24 units (n2) were chosen at random. The
list was created using Random number generator (refer sheet “Units being Sampled”). It has
been summarised below:

Units to be selected for second step of sampling


12 33 125 177
13 52 127 184
18 55 137 188
28 91 145 192
30 100 149 193
32 108 155 197

The summary of second step of Double sampling for the undecided lots are as below:

Lot No. of Defective No. of Defective Step 2 result


Step 1 result D1+D2
number units in Step 1 (D1) units in Step 2 (D2) (Accept|Reject)

7 1 Undecided 0 1 Accept
183 1 Undecided 1 2 Reject

3|Page
Quality Management Homework Assignment 1

We reject a lot if total number of defective units after evaluating both the samples is found
out to be greater than 1 (c2).
According to the decision to accept or reject a lot, the cost has been calculated as below
(detailed computation in attached excel file):

Cost of Cost of
Defects Cost of processing
Lot Final Sampling Sampling Total Cost
found while (for accepted lots)
numbers Results Step 1 Step 2 (EUR)
assembling (EUR)
(EUR) (EUR)
7 Accept 30 24 1 150 204
20 Reject 30 0 NA 0 30
26 Accept 30 0 12 1800 1830
34 Reject 30 0 NA 0 30
43 Reject 30 0 NA 0 30
62 Reject 30 0 NA 0 30
82 Accept 30 0 6 900 930
165 Reject 30 0 NA 0 30
183 Reject 30 24 NA 0 54
198 Reject 30 0 NA 0 30

Case 3 (Continuous Sampling):


For the continuous sampling plan we began sampling the units in our lots one by one in the
ascending order of the unit ids. We plotted the number of defects against the number of units
sampled along with the AQL and RQL lines with h1= -1.08; h2= 1.12; s=0.055 and identified
the decision point for each lot based on which of the lines get intersected first.
The results of the continuous sampling process are summarized in the table below. It has been
derived from the graph plotted in Figure 1 below the table. The decision point column refers
to the Number of units that had to be sampled in the respective lot for us to conclude an
Accept/Reject decision based on the defectives plot crossing the AQL or RQL.
Defects found Process
Decision Sampling while Cost (if Total
Lot Point Accept|Reject Cost assembling Accepted) Cost
7 20 Accept 20 2 300 320
20 48 Reject 48 NA 0 48
26 111 Accept 111 7 1050 1161
34 52 Reject 52 NA 0 52
43 5 Reject 5 NA 0 5
62 57 Accept 57 8 1200 1257
82 3 Reject 3 NA 0 3
165 67 Reject 67 NA 0 67
183 38 Accept 38 3 450 488
198 25 Reject 25 NA 0 25

4|Page
5|Page
Continuous sampling
20
Quality Management

15

7
20
26
10
34
43
62
82

Figure 1
165
183
5 198

Number of Defectives
AQL Plot
RQL Plot

0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 118 121 124 127 130 133 136 139 142 145 148 151 154 157 160 163 166 169 172 175 178 181 184 187 190 193 196 199

Number of units sampled

-5
Homework Assignment 1
Quality Management Homework Assignment 1

The column Defects found while assembling in the table above is calculated by subtracting the
number of defective units identified in the sampling process before reaching a accept decision
from the total number of defective pieces that existed in that lot. Only these unidentified
defective units in the lots incur the additional cost K2 based on our assumptions.
(2) Based on the cost calculations done for each sampling policy, the results are tabulated below
(refer sheet “Result Summary”):
Single Sampling Double Samplig Continuous
Lot number cost cost sampling cost
(EUR) (EUR) (EUR)
7 200 204 320
20 50 30 48
26 50 1830 1161
34 50 30 52
43 50 30 5
62 1550 30 1257
82 800 930 3
165 50 30 67
183 500 54 488
198 50 30 25
Total cost incurred 3350 3198 3426

We notice here that Double Sampling is the most economical policy in this case. However this
cannot be considered as a dominant logic because the cost computation depends on various
factors which may change from case to case basis.
The overall cost may change depending on the following factors:
i. Difference in the decision taken for various lots (accepted/rejected) by applying
various sampling policies
ii. The magnitude of cost of inspecting a lot and cost of processing a defective piece.
We notice here that even though some of the lots face the same final decision through each
policy, still there are some lots that are being treated differently. The difference in total cost
pertaining to any sampling varies greatly as per the acceptance or rejection of the lot. This
also further depends on the magnitude of cost involved in sampling procedure and cost of
correcting defective products. Since these factors varies according to circumstances so does
the overall cost incurred in different Sampling logics. Thus the results might vary if we would
have taken a different set of lots for inspection.

6|Page
Quality Management Homework Assignment 1

(3) Considering the AQL to be 5%, we get the following decision for each lot (refer sheet “Result
Summary”):

Hindsight
No. of Non- No. of Defective
Lot (AQL<=5% [i.e.
Defective units units
10 units])
7 198 2 Accept
20 190 10 Accept
26 188 12 Reject
34 186 14 Reject
43 184 16 Reject
62 190 10 Accept
82 194 6 Accept
165 182 18 Reject
183 196 4 Accept
198 192 8 Accept

A lot is rejected in case there are more than 10 (five percent) defective units.
On comparing the result with each sampling policy we observe the following table:
Lot Hindsight Single Double Continuous
(AQL<=5%) Sampling Sampling Sampling

7 Accept Accept Accept Accept


20 Accept Reject Reject Reject
26 Reject Reject Accept Accept
34 Reject Reject Reject Reject
43 Reject Reject Reject Reject
62 Accept Accept Reject Accept
82 Accept Accept Accept Reject
165 Reject Reject Reject Reject
183 Accept Accept Reject Accept
198 Accept Reject Reject Reject

We have the following observations for each lot :

 Lot 7, 20, 34, 43, 165, 198 : The same decision is taken through all the sampling
policies.

 Lot 26 : Though it should have been rejected but the decision is taken to accept the
lot in Double and continuous sampling.

 Lot 62 : Though it should have been accepted but the decision is taken to reject the
lot in Double sampling.

7|Page
Quality Management Homework Assignment 1

 Lot 82 : Though it should have been accepted but the decision is taken to reject the
lot in Continuous sampling.

 Lot 183 : Though it should have been accepted but the decision is taken to reject the
lot in Double sampling.

We notice in this case that the best results were obtained using Single sampling method. The
number of correct evaluations made in each policy have been summarised below:

 Single Sampling : 8 of the lots were correctly evaluated.


 Double Sampling : 5 of the lots were correctly evaluated.
 Continuous Sampling : 6 of the lots were correctly evaluated.

However this cannot be considered as a dominant logic because the effectiveness of sampling
logic will depend on various factors like the parameters (n, c, n1, c1, n2, c2 etc.) taken into
consideration for each type of sampling, the units being randomly selected, and the decided
AQL level. The actual defect percentage in the lots is one of the major influencers for
determining which sampling process would be best suited for the lot. Double sampling or its
extreme version continuous sampling is more effective when the given lot has very low or
very high defect percentage (i.e. <2% or >8 %) for lots with intermediate defect percentages
single sampling is the better choice. As these factors vary so could the final decision for each
lot.

8|Page

You might also like