This document outlines the development of an optimization tool for scheduling preventive maintenance and renewal projects of railway infrastructures. It describes the challenges faced by ProRail in scheduling €350M annually in renewal projects and €150M in maintenance projects. The tool aims to minimize costs by grouping related projects and balancing maintenance with renewal, optimizing when each component is serviced across track segments over time. A case study applying the tool to a 10km railway track found a 14% reduction in costs compared to the previous spreadsheet approach.
2. Maintenance scheduling- Farzad Pargar
Problem description
Solution approach
Introduction to the developed tool
The importance, contribution and benefits of the tool
Case study
How to implement the tool
OUTLINE
2
4. Maintenance scheduling- Farzad Pargar
4
DUTCH RAILWAY NETWORK
THE MOST INTENSELY USED RAILWAY NETWORK IN EUROPE
Figure 1. Dutch railway network and infrastructures.
5. Maintenance scheduling- Farzad Pargar
5
PRORAIL
INFRASTRUCTURE MANAGER FOR THE DUTCH RAILWAY NETWORK
Figure 2. The institutional triangle: key roles and business relations.
ProRail carries responsibility for new construction, maintenance, renewal,
capacity allocation, as well as traffic management.
6. Maintenance scheduling- Farzad Pargar
6
PRORAIL
THE CHALLENGES
• How to collaborate with its stakeholders to meet their mutual goals on getting
access to the available resources during rail operation and maintenance.
• Each year about EUR 350 million is involved in renewal projects and about
EUR 150 million is involved in maintenance projects.
• It is necessary to have a sophisticated tool to determine the optimal renewal
time and schedule maintenance projects.
8. Maintenance scheduling- Farzad Pargar
Title:
Preventive maintenance and renewal scheduling of railway infrastructures
Preventive maintenance is a set of activities that can be planned ahead
to prevent failures from occurring.
Scheduling is a decision-making process of allocating resources over
time to perform a collection of required tasks.
Setup time/cost, in general, can be defined as the time/cost required to
prepare the necessary resource (e.g., machines, people) to perform a
task (e.g., maintenance operation).
8
DEFINITIONS
PREVENTICE MAINTENANCE SCHEDULING
9. Maintenance scheduling- Farzad Pargar
PROBLEM DESCRIPTION
WHICH INFRASTRUCTURE?
9
Railway track
• The track assets of ProRail have a replacement value of about EUR 8 billion.
• Over 50% of the maintenance costs and 70% of the renewal costs are due to
the tracks.
10. Maintenance scheduling- Farzad Pargar
10
PROBLEM DESCRIPTION
TRACK AS A MULTI-UNIT SYSTEM
A multi-unit system includes several multi component systems in sequence.
Structure of a typical multi-unit system with several components.
13. Maintenance scheduling- Farzad Pargar
Track condition is not uniform in all locations and different locations has
different maintenance requirements. Thefore, we devide each section into
units.
In each unit the traffic, track conditions, and component types are all
uniform.
The downtime cost of such a system due to maintenance is high.
13
DIVIDING TRACK INTO UNITS (SEGMENTS)
A
B
14. Maintenance scheduling- Farzad Pargar
1. Budget determination
2. Long-term quality prediction
3. Project identification & definition (diagnosis)
4. Project prioritization and selection
5. Possession allocation and timetabling of track possession
6. Project combination
7. Short term maintenance and project scheduling (balancing available capacity)
8. Work evaluation and feedback loop
MAINTENANCE PLANNING OF RAILWAY
INFRASTRUCTURE
DIFFERENT DECISION PHASES
14
17. Maintenance scheduling- Farzad Pargar
17
MODEL FRAMEWORK
OBJECTIVE
Input data Output results
Optimization
tool
- Maintenance & replacement costs
- Setup costs
- Maintenance history
- Time interval between maintenance
projects
- Maintenance and replacement of
each component in different units
should be done in which period.
Our aim is to develop an integrated optimization tool which …
• Groups both maintenance and renewal projects simultaneously
• Finds the optimal balance between doing maintenance and renewal
By minimizing the cost of maintenance and renewal and their relevant setup
activities over planning horizon.
18. Maintenance scheduling- Farzad Pargar
18
OUTPUT RESULTS
HOW SOLUTIONS LOOK LIKE?
Units
Planning horizon
1 2 3
1
k
T
r1
m1 : maintenance of component 1
r1
m1
m2
m2
r1
r2
Maintenance history
r1 : renewal of component 1
r2 : renewal of component 2 m2 : maintenance of component 2
r2
……
…
…
…
20. Maintenance scheduling- Farzad Pargar
Maintenance activities are imperfect.
The latest possible times between maintenance/renewal activities for each
component (planning cycle) are known. These time intervals may not be fixed
and could have a decreasing order.
There is a maximum allowed number of times that maintenance on each
component can be performed.
20
MAIN ASSUMPTIONS
21. Maintenance scheduling- Farzad Pargar
21
TECHNIQUES TO REDUCE PM COSTS
WITHOUT REDUCING THE LEVEL OF MAINTENANCE
Grouping:
A technique to combine the executions of maintenance and replacement
activities.
Grouping allows savings on the preparatory works.
Balancing:
A technique to find a balance between maintenance and renewal.
Balancing allows to replace the component on its economic life rather than
technical life.
22. Maintenance scheduling- Farzad Pargar
22
BENEFITS OF GROUPING
EXAMPLE: 3 UNITS, 2 COMPONENTS, 3 TIME PERIODS
Costs of preparatory works
Replacement of
component 1
Maintenance of
component 2
System down time cost (Euro/period) 100 100
Preparation cost of equipment (Euro/period) 30 20
Installation cost of equipment (Euro/unit) 5 5
Here, we compare three scenarios for carrying out the replacement of
component 1 in units 1-3 and maintenance of component 2 in units 1-2.
23. Maintenance scheduling- Farzad Pargar23
BENEFITS OF BALANCING
EXAMPLE: 1 UNIT, 1 COMPONENT
When an asset ages, maintenance is required increasingly often.
The latest possible time for carrying out the next maintenance/renewal
relevant to the previous maintenance is known (planning cycle).
Wear stock
Time
Technical life : Renewal
: Maintenance
New
Planning horizon
Maintenance
threshold
PM cycle 1
24. Maintenance scheduling- Farzad Pargar24
BENEFITS OF BALANCING
EXAMPLE: 1 UNIT, 1 COMPONENT
When an asset ages, maintenance is required increasingly often.
The latest possible time for carrying out the next maintenance/renewal
relevant to the previous maintenance is known (planning cycle).
Wear stock
Time
: Renewal
: Maintenance
New
Maintenance
threshold
Planning horizon
Wear stock
Time
PM cycle 1
Economic life
30. Maintenance scheduling- Farzad Pargar
30
EXAMPLE
SEQUENCE OF MAINTENANCE AND RENEWAL PROJECTS WITHOUT OPTIMIZATION
Parameter Value
Components
Units
Time periods
Number of allowed maintenance
Planning cycles
I=2
K=3
T=16
H=2
𝑇𝑖ℎ
𝑙
=
3 1
2 2
Units
The sequence of maintenance and renewal projects without optimization.
31. Maintenance scheduling- Farzad Pargar
31
EXAMPLE
EFFECT OF PLANNING HORIZON (T) & DOWNTIME COST (𝑆𝐶 𝑑)
Optimal maintenance and replacement schedule with S𝐶 𝑑
= 500 and T=8.
Optimal maintenance and replacement schedule with S𝐶 𝑑
= 500 and T=16.Optimal maintenance and replacement schedule with 𝑆𝐶 𝑑
= 1000 and T=16.
33. Maintenance scheduling- Farzad Pargar
Decompose the problem into less complex sub-problems and then solve
them sequentially.
First, focus on optimal balancing for each component in each unit. The
grouping of projects is added later to the schedule.
33
HEURISTICS
DECOMPOSITION APPROACH
34. Maintenance scheduling- Farzad Pargar
34
CASE STUDY
RAILWAY TRACK
The European countries are reported to allocate 15-25 billion EUR annually
on maintenance and renewals for a railway system consisting of about 300
000 km of track giving an average of 70 000 EUR per km track and year.
The presented heuristic algorithms resulted in a solution with much lower
total cost in comparison with the spreadsheet based approach used in
practice; it resulted in 14% reduction in cost.
: I=3, K=10 (10 KM), T=40 (120 MONTHS)
35. Maintenance scheduling- Farzad Pargar
Developing a method that can be used in capacity management, maintenance
and renewal decision making process by considering conflict interests of
stakeholders
Tool’s applications:
Possession allocation and project combination
Compare different maintenance and renewal scenarios
It can be adapted for scheduling routine maintenance activities and be used
for other multi-unit systems
Tool’s users:
Technical specialists (vakdeskundige) in different regions and Plan
coordinators in the Asset Management department
The tool can be used by both infra manager and contractors for different
types of systems. 35
IMPORTANCE AND BENEFITS OF THE TOOL
TOOL’S APPLICATIONS
36. Maintenance scheduling- Farzad Pargar
A pilot study has been performed by following the above phases.
Developing a sophisticated tool/software out of the results of this
research would be beneficial for ProRail and other parties.
36
WHAT IS NEEDED TO BE DONE?
THE SOFTWARE DEVELOPMENT LIFE CYCLE
1.
2.
3.
4.
5.
37. Maintenance scheduling- Farzad Pargar
An optimization tool is proposed for maintenance planning of multi-unit systems.
The proposed tool reduce the costs of maintenance without reducing the amount
of maintenance by grouping maintenance and renewal projects and simultaneously
finding the optimal balance between doing maintenance and renewal.
The maintenance and renewal projects are scheduled in such a way that the sum
of infrastructure possession cost, machine setup costs and maintenance/renewal
process costs are minimized.
A good model is not a model in a paper or in a presentation but a model that is
in use.
37
CONCLUSIONS
38. Maintenance scheduling- Farzad Pargar
Pargar, F. (2015). A mathematical model for scheduling preventive
maintenance and renewal projects of infrastructures, Safety and
Reliability of Complex Engineered Systems, ISBN 9781138028791, pp.
993–1000.
Pargar, F., Kauppila, O., and Kujala, J. (2017). Integrated scheduling of
preventive maintenance and renewal projects for multi-unit systems with
grouping and balancing, Computer and Industrial Engineering, 110, pp.
43-58.
38
REFERENCES