Slide No.1 ● The Diesel Locomotive Shed at XXX, homes 181 locomotives. These locomotives visit the shed periodically for maintenance and repair. ● The shed operates on a 24x7, 365 days a year basis. Locomotives enter and leave the shed round the clock. ● The shed provides about 20 locomotives for working passenger carrying trains, which are scheduled as per a published timetable. Thus these locomotives leave and arrive from/to the shed at predetermined times of the day. ● The remaining locomotives are provided for freight operations which are more or less random in terms of arrivals/departures to/from the shed. ● The primary objectives of the shed are to ensure adequate Availability and Reliability of the Locomotives for working Trains. ● The facilities available for this purpose include maintenance bays, machinery, plant and equipment to carry out the maintenance, materials (spare parts) required for maintenance and trained manpower to carry out the maintenance. Railroad Industry-Reducing locomotive maintenance time
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Slide No.1
● The Diesel Locomotive Shed at XXX, homes 181 locomotives. These
locomotives visit the shed periodically for maintenance and repair.
● The shed operates on a 24x7, 365 days a year basis. Locomotives enter
and leave the shed round the clock.
● The shed provides about 20 locomotives for working passenger carrying
trains, which are scheduled as per a published timetable. Thus these
locomotives leave and arrive from/to the shed at predetermined times of
the day.
● The remaining locomotives are provided for freight operations which are
more or less random in terms of arrivals/departures to/from the shed.
● The primary objectives of the shed are to ensure adequate Availability and
Reliability of the Locomotives for working Trains.
● The facilities available for this purpose include maintenance bays,
machinery, plant and equipment to carry out the maintenance, materials
(spare parts) required for maintenance and trained manpower to carry out
the maintenance.
Railroad Industry-Reducing locomotive
maintenance time
Slide No.2
My Focus Dimension -Time I have chosen Time as the key dimension that I want to focus on. The time taken for a
locomotive to be serviced by the shed should be minimized. These could be broken into
● Flow Time i.e. the time taken to carry out the maintenance/repair work.
● Wait Timed being the time spent by the locomotive while it waits for maintenance
attention
and
● Locomotives Turned Away: The instances when a locomotive requiring maintenance is
not admitted into the shed owing to lack of capacity (customers turned away).
The trade offs could be across multiple pairs of the four dimensions of the efficiency frontier, for
example Time v/s Cost, Time v/s Quality, Quality v/s Cost Time v/s Variety etc.
From the perspective of the Shed management, the key tradeoff is between Quality and Time.
I will be focussing on this pair of dimensions and attempt to strike an optimum balance between
Quality and Time.
Key Trade Offs- Quality V/s Time
The purpose of this project is to help the shed improve upon the availability
metric by minimizing the time taken to carry out the maintenance and repair
activities.
Slide No.3
Sensing the problem
The key performance measures in this context as related to the four dimensions of performance
are:
1. Time: The time taken to carry out the maintenance/repair work determines the availability of
the Locomotive (flow time). Lesser the time taken, greater the availability for working trains.
Other parameters would be the time spent by the locomotive while it waits for maintenance
attention (wait time), and the instances when a locomotive requiring maintenance is not
admitted into the shed owing to lack of capacity (customers turned away).
2. Quality: The quality of maintenance determines the reliability of the Locomotive. It also affects
the availability since a failed locomotive is not available to work a train. Number of locomotive
failures per million kilometers run measures this parameter (defect rate).
3. Cost: The cost of staff and materials used to carry out the maintenance and repair activities is
an important performance metric for the shed. The relevant parameters would be Maintenance
cost per Locomotive, Staff required per Locomotive, Idle time of staff etc.
4. Variety: The shed turns out single locomotives, Multiple Units (MUs) i.e. two or more
locomotives coupled together. The customer’s requirements are dynamic and the shed is
required to match the customer’s requirement with respect to singles and MUs
Slide No.4
Objective- What would Success
Look Like?
Based on my interaction with the users (customers) and the shed
management, the vision of success would be as follows:
1. The shed will improve the Availability of its Locomotives by 1%
(Quantity/Throughput).
2. The shed will be able to utilize its resources optimally to attend to the
maintenance and repair requirements of the Locomotives (Efficiency)
a. Reduce material cost by 10%
b. Reduce Labour Content by 10%
Slide No.5
Problem Formulation How can we reduce the time taken for maintenance?
Alternative#1:How can we reduce the maintenance time for various
activities? (Narrow scope)
A: By reducing time taken for each maintenance activity and reducing
waiting times.
Alternative#2: Why should we reduce the time taken for
maintenance? (Broaden scope)
A: The locomotives can earn more on line
Alternative#3: Q: How can the locomotives earn more on line?
(Narrow scope)
A: By visiting the shed less frequently and by taking minimum time in
each visit.
Alternative#4: Q: Why would locomotives visit the shed less
frequently and why would they require minimum time each visit?
(Broaden scope)
A: If the locomotives are more reliable
Slide No.6
•I have defined the KPI here as the Avg. No. Of locomotives inside the shed per day. The lower this number, the
better the performance.
•This is a function of the Avg. Time spent by a locomotive and the Avg. No. Of locomotives for each type of
maintenance/repair activities.
•The sliders on the right side are used to know the impact of improvement in each parameter
KPI Tree: Number of Locomotives in Shed
% Improved
Locomotives
Saved
Slide No.7
Explanatory Notes and Inferences
• The Locomotives visit the shed for Monthly (30 days),
90/180/270 Day schedules, Yearly Schedules, Three Yearly
Schedules, Six Yearly Schedules and Out of Course
(Unscheduled) repairs.
• Based on the historical data available and the KPI tree, the “M”
(Monthly) schedule has the biggest impact.
• I have therefore prepared the Process Flow Chart for the “M”
schedule.
• I Have not provided a slide for the customer’s perspective, since
the flow unit here is a locomotive.
• The first four Slides pertain to COP Assignment#1. I have
continued COP Assignment#2 in the same file for continuity.
• Activity#6 in the process flow chart i.e. The Drop Pit appears to
be the bottleneck.
Slide No.8
S.No. Activity (time in minutes/%
Attended)
S.No. Activity (time in minutes/% Attended)
1 Placement in Bay (30/100%) 5 Mechanical Repairs (180/30%)
2 Mechanical Maintenance
(240/100%)
6 Drop Pit (180/40%)
3 Lab Test of Oils and Water
(60/100%)
7 Electrical Repairs (240/10%)
4 Electrical Maintenance
(320/100%)
8 Washing filling of Supplies and despatch
(60/100%)
Locomotive Monthly Schedule
Process Flow Diagram
1
2
3
4
5
6
7
8
Locomotive
released
from
Operations
for Monthly
Schedule
Slide No.9
Peer feedback of COP#2
peer 5 → “The presentation is confusing; the problem formulation is like a cat
running to catch its tail: no consistent answer is given The KPI tree does't
lead to costs, revenues, profits, it's closer to a list of number of locomotives
(why so many decimal digits?) The customer perspective is missing even if
it could be done. EG: what does it mean in terms of waiting times, possible
delays, breakage during the trip, ..... whatever... The process flow diagram
is the only thing closer to the requests, but it's not clear what number
represents and why it was chosen the total process and not only a
significant part. In addition to this, no waiting/triangle is explained ...”
Based on the feedback from COP#2, I understand that I need to clarify
certain to help my evaluators have a better perspective of the issue (In spite
of having included a slide titled explanatory Notes”!). The points raised by
one of the Peer evaluators are reproduced below:
Clarification on my Problem Formulation:
I started with “How can we reduce the time taken for maintenance?”
I have then attempted to alternatively broaden and narrow the scope to help
view the problem from different perspectives.
Slide No.10
Peer feedback of COP#2 About the KPI Tree: I have not shown costs here because, in my
organization, there is no “cost” assigned to the locomotive in terms of, say,
$/hour. However, it is intuitive that the lesser the number of locomotives
inside the Shed the better it is. To that end, I have collected data of each
locomotive entering and leaving shed over a period of nine months from 1st
Jan’13 to 30th Sep 13.
The KPI Tree shows the impact of the Number of Locomotives visiting the
shed for each type of Maintenance/Repair and the Time taken to carry out
the maintenance/repair. Thus, more locomotives would be inside the shed if
either more locomotives visit the shed and/or if more time is taken to service
each locomotive. This is what the KPI Tree shows. I have made this Tree in
Excel and built sliders by the side of each parameter. This slider can be
adjusted from 0 to 100%. So if I assume a 10% improvement in service time
for the Six yearly Schedule, I can see the effect of such an impact on the
KPI by merely moving the relevant slider.
I have edited the KPI tree to limit the numbers to two decimal places.
Slide No.11
The Process Flow Diagram:
I have mentioned in Slide 8 that I have chosen the M schedule since it
appeared to have the maximum impact on the KPI. The process flow was
thus only for the M schedule. The waiting triangles indicate the locomotives
may have to wait to get attended at the subsequent operation in case the
work station is busy attending to another locomotive. The numbers are
explained in the Legend to the diagram, e.g., 180/30% indicates that the
operation requires 180 minutes and 30% of the locomotives go through this
operation.
Peer feedback of COP#2
My submission for
COP#3 Starts from
the next slide
Slide No.12
In COP#2, I had identified the Monthly Schedule as the potential
activity which would “give the biggest bang-for-the-buck” as far as
the KPI is concerned.
I have studied the process of the monthly schedule being carried out
on the locomotive.
I have however analysed the data and found that:
1. There was little scope for improvement in the M schedule. I could
not find any significant presence of any of the seven sources of
waste.
2. On the other hand, I found that the variability is more in Y and 3Y
schedules.
3. There is evidence of over processing in the Y,3Y and 6Y
schedules. This discovery was serendipitous, i.e. My finding was
unexpected.
Collecting Primary Data
Slide No.13
Identifying Sources of Waste
Of the seven sources of waste I found that Over processing is a major
contributor to inefficiency with respect to the Y, 3Y and 6Y
schedules.
As per the OEM of the Locomotive, the 6Y Schedule is to be carried
out after the Locomotive produces 23,000 Megawatt hours of power.
The lower schedules (3Y and Y) are also proportionately spaced.
Analysis of data shows that the freight locomotives generate this
amount of power in about 9 years.
So all activities which are presently being carried out in Y6 can be
postponed to Y9.
This would cascade to Y3 and Y schedules also.
I have checked the implication from the technical aspect also and am
satisfied that such an extension is justified.
Time taken for Maintenance/Repair
Schedule Type
Number of Locos
attended over the
period
Average time
spent in shed
per loco (days)
Average
Locomotives
in Shed
MONTHLY 1512 1.22 6.74
90/180/270 Days 372 1.84 2.50
Y 61 7.30 1.62
Y3 22 21.68 1.74
Y6 24 37.96 3.32
OUT OF
COURSE 138 3.23 1.63
Grand Total 2129 2.26 17.56
Primary Data
The data pertains to the period from 1st Jan 2013 to 30th Sep 2013.
This was compiled from the available database maintained in the
shed. Side No.14
Slide No.15
Process Variability
X Axis: Month (1=Jan) Y Axis: Time taken in Days
0.5
0.75
1
1.25
1.5
1.75
2
1 2 3 4 5 6 7 8 9
Avg Mly Time
LCL
UCL
0
2
4
6
8
10
12
14
1 2 3 4 5 6 7 8 9
Avg Yly Time
LCL
UCL
-1
9
19
29
39
49
59
69
79
1 2 3 4 5 6 7 8
Avg 3 Yly Time
LCL
UCL
20
25
30
35
40
45
50
1 2 3 4 5 6 7 8
Avg 6 Yly Time
LCL
UCL
Slide No.16
Demand forecast after review of periodicity of the
Y,3Y and 6Y schedules
Schedule Existing
Periodicity
Proposed
Periodicity
Existing
Demand
per Year
Demand
per Year
based on
proposed
periodicity
Y 1 year 1.5 years 80
locomotives
71
locomotives
3Y 3 years 4.5 years 54
locomotives
36
locomotives
6Y 6 years 9 years 27
locomotives
18
locomotives
Note: There are 161 freight locomotives. Demand is
calculated as follows:
6Y Existing: 161/6= 27 locomotives
6Y Proposed: 161/9= 18 locomotives
Etc.
Slide No.17
Inflexibilities Noticed
I have found evidence of Volume inflexibility: The capacity
which is fixed, is not able to adjust to the demand which