José Antonio Moreno Highly automated railway wheel set maintenance workshop supported by simulation

Danobat S.Coop

HIGHLY AUTOMATED RAILWAY WHEEL SET MAINTENANCE

WORKSHOP SUPPORTED BY SIMULATION

José Antonio Moreno

MSc Industrial Engineer
Danobat S.Coop
Itziar Ricondo Ander Azkarate
PhD Industrial Engineer PhD Industrial Engineer
IK4-Ideko S.Coop IK4-Ideko S.Coop

Summary
The maintenance of wheel sets is a challenging task regarding process definition and part routing, due
to the increasing variety and need of flexibility. Heavy Haul Operators have invested considerable time
and effort learning how to run the world's most efficient ore car maintenance wheel shops. Main
contractors of this kind of wheel shop (usually Machine Tool Manufacturers) have transformed the way
Heavy Haul Operators perform its maintenance task from manual workshops to more efficient and
automated wheel shops. This paper will present a simulation-based optimal design and analysis method
for designing & optimizing a railway wheel set maintenance workshop. A case study for the above
approach will be applied using computer application software for modelling, simulating, analyzing,
visualizing and optimizing maintenance workshop and its processes and flows.

1. INTRODUCTION task from manual workshops to more efficient

and automated wheel shops. The demand of
The maintenance performed by a company has
wheel shop automation is growing, connecting
a significant impact on its business [1]. Train
different equipment to each other and taking
operators make every effort to maximize fleet
preventive maintenance decisions rather than
availability and minimize related maintenance
corrective ones, extending wheel set life and
costs. The wheel set is the part of the train that
decreasing maintenance costs
carries the heaviest portion of the load, and high
axle load, high speeds, shocks and extreme Actually an ore car maintenance wheel shop is
weather conditions make them sensitive to a highly automated manufacturing system
damage and wear and tear. considering multi-stage process plans [3] [4].
In terms of Rolling Stock maintenance costs, This type of wheel shop provides an answer to
wheel set life cycle cost is one of the most growing demand for fully automated plants to
relevant operational efforts [2]. This occurs to a perform maintenance on wheel sets (mainly
greater extent with Heavy Haul Operators due mining or freight), either located near the mining
to the fact that wheel sets are subject to higher site or in remote areas which therefore have a
wear and damage. shortage of nearby labour. Disassembly, repair,
reassembly and final inspection of all
For that reason, Heavy Haul Operators have
components could be performed in a fully
invested considerable time and effort learning
automated way on these installations.
how to run the world's most efficient ore car
maintenance wheel shop. Main contractors of We are entering a new era of manufacturing,
this kind of wheel shop (usually Machine Tool one in which computers and machines are
Manufacturers) have transformed the way coming together to perform automated actions.
Heavy Haul Operators perform its maintenance More impressively, they are able to do this with

AusRAIL PLUS 2017

21-23 November, Brisbane
José Antonio Moreno Highly automated railway wheel set maintenance workshop supported by simulation
Danobat S.Coop

less and less human involvement, learning from tools have been used helping decision making
their mistakes and teaching themselves to process.
constantly perform better.

The world of The Digital Factory is helping main

suppliers of ore car maintenance wheel shops 3. COMPLEXITY IN SHOP AUTOMATION
to design and simulate different scenarios in Manufacturing paradigms are challenged to
order to align customer’s requirements with the constantly improve production systems in terms
complexity in shop automation. of flexibility, reliability and responsiveness to
satisfy customer demands for products with
better features. To meet production targets of
2. DIGITAL FACTORY complex products with higher quality
requirements and reduced time to market, the
Digital Factory represents a virtual picture of a
manufacturing industry is commissioning highly
real production [5]. It is the environment
automated production systems, numerous sub-
integrated by computer and information
systems of various natures, including machining
technologies, in which the reality is replaced by
and processing systems, material handling
virtual models. Digital Factory tools include
devices and material storage and retrieval units.
CAD/CAM tools, part process and production
Ore car workshops for wheel set maintenance
simulation tools and their integration to real
are designed under this complexity. This
systems.
section explains which are the main elements
Simulation of the complete material and and sources of complexity for these advanced
information flow is recognized as a key maintenance shops.
component of the Digital Factory in the industry.
3.1 Shop automation
Among different simulation approaches,
discrete event simulation is well suited to Increasing efficiency requirements are
modeling manufacturing systems [6]. At motivating the introduction of automation to
industrial level, the simulation of manufacturing facilitate the design and operation of production
systems is performed using commercial and maintenance plants [7]. Automation needs
software, rather than through a purpose built to be addressed in all levels, from
application. These commercial tools include machine/device level to business systems
application oriented objects or functions that (ERP), suited to customer requirements.
increase ease of use and reduce model building Automation of the whole shop includes the
time. As an example, some commercial equipment (machines, auxiliary equipment,
simulation packages are: Tecnomatix Plant robots, manipulators, etc.), as well as the
Simulation, Rockwell Arena, Anylogic or Simio. monitoring and control of production.

The use of Digital Manufacturing tools allows 3.2 Production control system
making faster, more reliable and smarter The Control System is responsible for part
decisions in the design and development of monitoring, part control, traceability and safety
turnkey projects for wheel set maintenance functionalities. Part control and routing is set
workshops and new railway manufacturing dynamically as a result of both manual and
plants. Modelling and virtual capabilities are automatic inspection results, as well as
enhanced with optimisation capabilities, so the machining parameters of certain stations (e.g.
efficiency of multi-stage production facilities for wheel boring). Traceability may require the
wheels, axles, wheelsets and bogies can be integration to external enterprise systems (ERP,
increased. MES, etc.).
A highly automated workshop to overhaul iron 3.3 Complexity of the maintenance process
ore wagons operating on BHP´s Iron Ore and shops
Pilbara rail network is now ready to open. At The maintenance of wheel sets is a challenging
different stages of the design phase simulation task regarding process definition and part

AusRAIL PLUS 2017

21-23 November, Brisbane
José Antonio Moreno Highly automated railway wheel set maintenance workshop supported by simulation
Danobat S.Coop

routing, due to the increasing variety and needs automatically routes the wheel set to the
of flexibility. This directly impacts the design of wheel set demounting area.
the layout. The sources of complexity and
- Production control and strategies: The need
variation are:
of efficiency is moving facility and
- Maintenance tasks, damage types and equipment providers to the introduction of
damage identification: The wheel sets may lean techniques. For example, pull control
have several damage types, which can be technologies oriented to production
identified at different inspection stages in synchronization and buffer reduction
the process and may impact the repair strategies are considered in real
needs of the wheel sets. This may also workshops.
cause variable repair times at certain
- Expansion needs: In the design of large
stations, depending on the damages. This
facilities, flexibility of the shop to future
also happens in the case of ore cars and
production expectations also need to be
bogies, with a large list of damage types.
considered.
The definition of a complete maintenance
event tree, defining the probabilities of 3.4 Integrated approach and required
occurrence/detection of damages and competences
corresponding repair time may help in
understanding the complete damage and The design, execution and commissioning of
repair scenarios. these maintenance shops is a complex process.
Therefore, those projects require an integrated
- Component variety and production mix: approach as well as multi-disciplinary
Maintenance shops have to be prepared to capabilities and competences, which go beyond
handle with an increasing number of wheel pure technological complexity. Moreover,
set types, which may have production and suppliers or Main Contractors of these shops
logistic implications for wheel sets, wheels, are expected to have proven expertise in the
axles and bearings. On the other hand, as whole scope and phases of wheel shop
wheel set types may have different projects, which include concept, design,
requirements, the production mix is also execution, commissioning and operation phase.
something to be managed, analyzing the In our experience, these are the required
impact of different mixes. competences for the successful execution of
those complex shop projects: process
- Maintenance strategies and decision
knowledge, equipment design and/or selection
making: Wheel sets require different
skills, automation, project management
maintenance strategies, attending to the
capability and the provision of services along
state of the wheel and axle. Reprofiling and
the whole project life cycle. This approach is
remount are two main strategies, as a result
represented in Figure 1.
of the inspections (wheel profile, back to
back, rim thickness, journal diameter, etc.)
during the process. Moreover, bearing and
end cap replacement may also be required,
even in the case that the wheel set does not
need to be demounted. If measurements
are outside the set parameters, the wheel
set is flagged to be demounted, otherwise it
is re-profiled. This decision may be made by
the operator (manual or visual inspection)
or automatically (in case of automatic
inspection). In fully automated wheel shops,
the decision of wheel set demount is
passed to the Control System, which

AusRAIL PLUS 2017

21-23 November, Brisbane
José Antonio Moreno Highly automated railway wheel set maintenance workshop supported by simulation
Danobat S.Coop

operational support, process and production

simulation.

4. SIMULATION-BASED DESIGN OF A
WHEEL SET MAINTENANCE FACILITY
The complex task of the design of an automated
wheel shop can be supported by the use of a
simulation tool, as a complementary tool to
current design tools and procedures. This
section explains a method to consider
Figure 1: Competences for maintenance simulation in the design process, in order to
shop suppliers develop more efficient and productive wheel
shops.
Deep knowledge of the manufacturing
processes itself, closely linked with customer
needs, is one of these key competences. Main 4.1 Simulation based design method for
contractors have usually developed their own wheel set maintenance shop
The method identifies four main steps in the
tailor-made range of machines and equipment
development of simulation models, as a
aimed at the needs of the rail sector, although
supporting tool during the design and
they also have the skills for the selection and
commissioning of wheel set maintenance
procurement of equipment from third parties
shops.
too.

Process knowledge is enhanced by the use of

a) Definition of the objective and scope of
digital tools, which increases robustness of the the model
solutions. Production simulation technologies The first step in the simulation method is to
are specific tools that allow the anticipation of establish the objective and scope of the model,
problems and provide collaborative solutions to as well as the level of detail to be considered,
the processes and production flow more which will directly impact the programming effort
effectively. and time. Although this stage can be divided
Automation is a critical competence in into several steps, our experience has taught us
advanced wheel set maintenance shops, that it is important to clearly define the
directly impacting productivity, quality, safety objectives of the model, the scope and proper
and overall shop efficiency. The final degree of abstraction level and to define an input and
automation is based on customer requirements. assumptions document, where the input data
and any other assumption or simplification are
There is no doubt for the need of project clearly stated.
management capabilities to deliver those
complex projects in quality, time and costs. The The main objective of the simulation model is to
use of project management tools (e.g. PMP® validate that the proposed design fulfils the
certification), with special attention to the final required throughput of the maintenance wheel
acceptance stages, commissioning and training shop, but some other sub-objectives from a
are necessary to ensure an efficient supplier- design point of view are: identify bottlenecks,
customer knowledge transfer. define required buffers, validate the production
control strategy or define priority rules of
Finally, these maintenance shops projects are material handling systems.
supported by a whole range of engineering
services during the life cycle, such as Technical The input and assumptions document is a
Support Assistance, customised maintenance, required milestone, before the beginning of
model building. In those complex wheel shops,
both production equipment and material

AusRAIL PLUS 2017

21-23 November, Brisbane
José Antonio Moreno Highly automated railway wheel set maintenance workshop supported by simulation
Danobat S.Coop

handling systems are critical for performance, the design process and optimization of
so the scope includes the machines, robots, certain parameters.
operators, gantries or AGVs required to
Due to the iterative nature of the design
understand and validate the functioning of the
process, model building and testing can be
system.
executed several times, as design evolves
towards final and optimal solution.

b) Development of the model

The development of the model is supported by
some programming tools, usually a discrete d) Simulation results
The output of the simulation should cover
event simulation commercial program. The
previously defined objectives and supports the
advantage of commercial programs is that they
validation of the final design of the system.
offer some predefined objects (machines,
Therefore, the result of the simulation is well
buffers, robots) and visual functionalities that
documented, including the inputs and
make modelling easier, although they also
assumptions and results of the performed tests.
provide programming capabilities to adapt
these general objects to customized project
needs.

The development of the model can be divided

into:

1. The introduction of the elements (machines,

handling elements, abstract objects such as
production program or calendars) into a frame
Figure 2: Simulation example
or even the layout.
Typical outputs of the model are: achieved
2. Modelling of the input data and modelling of throughput of the shop, number and use rates
the logic of the system. It is important to of equipment (including material handling
consider that the design is an iterative process, systems), number of parts in the system (WIP
so flexibility to re-adapt the model to new design level), labour requirements, bottleneck analysis
layouts and logics has to be planned in advance and different scenario analysis.
in the programming approach.
It is important also to think about the lifecycle of
c) Model running and test definition the model. For example, the simulation model
After model building, the resulting model is
may also be used during commissioning, or
internally validated, before starting the
even during the operation stage of the shop.
simulation of a set of tests that usually cover:

 the performance of the shop at 4.2 Benefits in a real case

standard conditions, in order to validate The explained simulation approach has been
the design proposal for production implemented in several cases of wheel shops in
requirements. Australia. In those cases, the Spanish machine
tool company Danobat S.Coop has been
 the performance of the shop to more
responsible for the wheel shop maintenance
stringent conditions or scenarios, such
equipment and the control of this equipment.
as an increasing number of failed wheel
One of the cases has been a new ore car repair
sets, modifications of the production
shop in Mooka for BHP Billiton Iron Ore. For this
mix or changes in the production
project, the simulation has been used during the
program.
design, in order to validate the design of the
 other design configurations in order to whole production system (process equipment
compare different alternatives during and production logic). The scope of the model
has been the ore car repair shop, comprising a

AusRAIL PLUS 2017

21-23 November, Brisbane
José Antonio Moreno Highly automated railway wheel set maintenance workshop supported by simulation
Danobat S.Coop

main production line, a bogie shop, a structural production requirements. Simulation allows the
shop and a wheel shop. The objective of the validation of the design and gives numerical
simulation has been the validation of the design, outputs of the main critical production decision
from preliminary to detailed design of the variables. The visual capabilities also enhance
solution, assessing that production the understanding. These benefits make
requirements were achieved, as well as testing simulation a very powerful tool to support the
the solution in other scenarios. design and achieve more efficient ore car wheel
set maintenance shops.
The simulation has been a very useful tool for
the development, understanding and validation
of the solution. For example, there are some REFERENCES
synchronization points in this repair shop, when [1] Rao B. Handbook of condition monitoring.
the ore car bogies are separated from the ore Elsevier. 1996.
car body and then when they are mounted again [2] Gallo S. et al. A new wheelset maintenance
in the process. As repairs to the bodies, bogies concept. World Congress on Railway Research
and wheel sets during the demounting and 2013.
mounting of bodies and bogies may vary, the
simulation has allowed checking and visually [3] Brymer B. Optimising Ore Car performance
validating the complete solution and also through design, construction and condition
enhances the understanding and based maintenance. International Heavy Haul
communication within the design time as well as Association 2015.
between the contractor and the customer. [4] Norris C. Automation in an ore car workshop.
Another complex task has been the modelling International Heavy Haul Association 2015.
of the diverse damage an ore car/bogie/wheel
set may have, which has been done with the [5] Kuhn W. Digital Factory - Simulation
use of a maintenance event tree, modelling the Enhancing the Product and Production
probabilities of occurrence, and repair times Engineering Process. Simulation Conference,
and repair stations. 2006.

The main benefit of the simulation is that it [6] Mourtzis D, Doukas M, Bernidaki D.
allows the validation of the design and gives Simulation in Manufacturing: Review and
numerical outputs based on the design input Challenges. Procedia CIRP. 2014;25:213–29.
data and assumptions. Moreover, due to its
[7] Groover MP. Automation, production
visual capabilities, these “living layouts”
systems, and computer-integrated
enhance the understanding of the operation
manufacturing. Harlow: Pearson Education;
logic of complex shops, since the user can see
2016.
how the shop is performing at every time.

5. CONCLUSIONS
This paper has shown how simulation can
support in the design and execution of efficient
wheel set maintenance shops. The design of
these wheel set maintenance shops is a
complex and challenging task, due to a variety
of maintenance tasks, product mix, part routing
and high number of equipment and automation
involved in the whole solution. It is hard to
consider all these complex issues in the design,
without the assistance of a simulation tool to
evaluate different design alternatives against

AusRAIL PLUS 2017

21-23 November, Brisbane