International Human Resource ManagementFull description
Description of Problem
Objectives of this workshop The rejections of paper manufactured by a company exceeds 30%. The magnitude of the problem has increased recently. Problem is observed at the end of the paper production process where the entire paper produced is tested for its quality.
Scope the problem Black Spots
Others
Thin Spots
Accepted product
Highest contribution to rejections Most serious
Rejection 36%
Product “A” 50%
Rs18 crore per annum
1% rejection corresponds to a revenue loss of Rs 50
Holes
Drawn to scale HOLES ARE THE MAJOR CONTRIBUTORS TO REJECTION. PROBLEM IS SCOPED TO: REJECTION OF PAPER CAUSED DUE TO PRESENCE OF HOLES
pet theory of the personnel in the plant was: “Roller Number 1 is root cause” picks up fibre or something and leaves a hole” Capex of Rs.2 crore would be required to replace the roller with a larger one.
We take a systematic problem solving approach, that uses: • only data and no opinion • strict logical analysis • experimental confirmation
THE DATA OF %DEFECTS OVER A PERIOD OF TIME INDICATED THAT THE PROBLEM IS OF VARIATION TYPE.
Is/Is-Not Matrix: Data Collection Format
Problem 1. What 2. Where 3. When 4. Quantity
Is Present in Problem
Is Absent in Problem
Data Obtained on “What” Problem
Present
Absent
What is the unit with defectPaper
Not applicable
What is the defect on the unit (this is the defect under study)
Holes
Not applicable
What other defects are present on the defective unit
None
Not applicable
What other defects are absent on the defective unit
Not applicable
Profile, Thickness, GSM, Porosity, Yellow is where etc.. data is available in this case
The Holes in the Paper
Hole Hole
The paper with holes does not have any other problems
Data Obtained on “Where” Problem Where – on which part of defective unit
Present
Absent
Anywhere across machine profile
Not present at any specific location
Hole Detector
Roller Number 1
Hole
ht di w C/ M
View across width of machine
Red: Medium size hole Blue: Large size hole WIDTH OF MACHINE
The Holes in the Paper Where is present? Problem Where – on which part of defective piece
Present
Absent
Anywhere across machine profile
Not present at any specific location
Absence of specific location, and presence of random spread suggests that the holes are not machine related, but related to something that contacts machine anywhere randomly
Roller Number 1
Hole
ht di w C/ M
Data Obtained on “Where”: Problem
Present
Absent
Where - on which part of defective piece Anywhere across Not present at any machine profile specific location Where – on which product variety
Imported RM, More hardwood.
Indian RM, Less hardwood.
Where – after which operation, and not before which operation
Cannot measure
Cannot measure
Where – on which machine or line
Only one m/c available
Only one m/c available
Where – in which plant/ area
Not checked
Not checked
Input feed to the process is randomly distributed
Data obtained in “where” indicates that: les do not appear to be correlated with machine and oles appear to be correlated with somethingCorroboration: Holes are randomly distributed across the process.
correlated with type of input into process
Detailed Problem Specs Sheet for “When” Problem
Present Absent
When – draw time line and show occurrences of defect When – in which shift and which part of the shift (hr/min) When – in which month; in which season When – in which year
Yellow is where data is available in this case
Detailed Problem Specs Sheet for “When” Problem
Present Absent
When – draw time line No clear pattern, and show occurrences of trend or defect periodicity When – in which shift and which part of the shift (hr/min)
No pattern visible
When – in which month; in which season When – in which year
Yellow is where data is available in this case
Data on “When” – Time line TREND TR-5000 ULMA % REJ ECTION 60 55 50
1
45 40 35 30 Series1
25 20 15 10 5 0
No clear pattern no build ups, no periodicity TREND TR-5000 ULMA %REJ ECTI ON
60 55 50
2
45 40 35 30 25 20 15 10 5 0
TREND TR-5000 ULMA % REJEC TION 60 55
3
50 45 40 35 30 25 20 15 10 5 0
Today
None of the recorded machine parameters and none of the recorded input parameters correlate with the ups and downs in the rejection percentage on the time line, when compared for the same grade of material.
Detailed Problem Specs Sheet for “Quantity” Problem
Present Absent
Quantity – How much (how deep/wide…) are the defects Quantity – How many defective pieces in the batch Quantity – How many defects on defective piece Yellow is where data is available in this case
Data Obtained for “Quantity”
Problem
Present
Quantity – How much (how Approx circular holes, which are deep/wide…) are the most common defects Quantity – How many defective pieces in the batch (roll) Quantity – How many defects on defective piece
Many
Absent Not steaks, tears,
Not zero
Physically search for “mura” (inconsistency) in input material
Found no contamination in input raw materials
Found foreign bodies in head box input!
Found foreign bodies in fresh water!
Found no contamination in chemicals and other streams
Found no contamination in recycle water!
Hypothesis: Foreign bodies from fresh water lead to holes
Foreign particles from fresh water are primary chronic causes of holes, resulting in rejections Machine parts, including “roller number 1” aggravate the problem, and are causes of instability problem (cyclic trend), resulting in rejections.
Countermeasures
•Reduce contamination in fresh water
This is a sporadic problem of holes in shape of cracks
This is a instability problem. Not to be confused with chronic problem. Cause for this is different, and known.
Difference between Sporadic and Shift
Sporad ic Shift We are solving this
Now the real challenge is on! 1. Completing the countermeasures 2. Sustaining the gains 3. Preventing and immediately correcting sporadic problems Can you meet it?
Should become redundant
Learning’s 8D approach makes no assumptions. It starts with data, from scratch on blank slate. uses logic to narrow down the area of search for the root cause. It does not require only technical expertise. It also requires reliable data. It obtains data by asking a series of questions. The questions are in strict sequence.