ETH Library

A Process to Enable the

Automation of Road Asset
Management

Conference Paper

Author(s):
Adey, Bryan T.

Publication date:
2017-06-29

Permanent link:
[Link]

Rights / license:
In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection.
For more information, please consult the Terms of use.
 Research Collection

Conference Paper

A PROCESS TO ENABLE THE AUTOMATION OF ROAD ASSET

MANAGEMENT

Author(s):
Adey, Bryan T.

Publication Date:
2017-06-29

Rights / License:
In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more
information please consult the Terms of use.

ETH Library
Zurich, June 29th-30th, 2017
2nd International Symposium on Infrastructure Asset
Management – SIAM 2017

A PROCESS TO ENABLE THE AUTOMATION OF ROAD ASSET MANAGEMENT

B.T. Adey1*

1 Institute of Construction and Infrastructure Management, Swiss Federal Institute of

Technology, Zürich, Switzerland
*Corresponding author; e-mail: adey@[Link]

Abstract
In this paper, a complete high level process to enable the automation of road asset
management is presented. It should be used by road asset managers who are
establishing their own more detailed process to ensure that all involved stakeholders
agree as to the activities involved in the management of road assets and the reasons
why. The use of this process to develop more detailed processes will help ensure that
the further efforts to automate all or part of road asset management will be useful. The
proposed process includes eight basic sub processes.

Keywords:
Framework; road asset management; automation, net-benefit

1 INTRODUCTION
1.1 General
As the field of road asset management becomes increasingly professional, increasing numbers of
researchers and software developers are attempting to automate some or all of the process. Some
examples for frameworks and processes to help encourage this are [1], [2], [3]. Some examples of tools
for different types of road assets can be found in [4], [5], [6]. Many of these efforts, however, fall short
of expectations with respect to automation, and result in large amounts of money being spent on
software that it is not used, not used for long, or used at far below the capabilities for which it was
intended. This happens in many cases because road asset managers lack a complete process, upon
which they agree, that is to be used for the management of road assets. Incomplete frameworks and
processes are useful in certain situations, but often fail when implemented in practice due to the real
world complexity of decision making. In this paper, a complete high level process is presented, to be
used to enable the automation of road asset management. It should be used by road asset managers
who are establishing their own more detailed process to ensure that all involved stakeholders agree as
to the sub processes involved and the reasons why. The use of this process to develop more detailed
processes will help ensure that the further efforts to automate all or part of road asset management will
be useful. The remainder of the introduction includes a definition of road assets and road asset
management.

1.2 Road assets

Road assets consist of the fixed physical assets 1 used to ensure the functioning of the road network.
The life cycle of a road asset is illustrated in Figure 1, where an asset, at each moment in time, is 1)

1The use of the word asset in all definitions is to allow it to be replaced by any term that refers to
physical infrastructure that is composed of smaller parts. For example, a network is comprised of
objects, objects are comprised of elements, and elements are comprised of segments. In such a case,

1
 SIAM 2017

being constructed, 2) being used, 3) being operated, 4) being maintained, 5) being developed, or 6)
being demolished. Where an asset is in its life cycle depends on its ability to provide an adequate service
level and the required service level. Both of which change over time. Provided service levels change
due to deterioration of the assets. Required service levels change due to changing needs for the assets.
This is illustrated in Figure 2.

Use

Operate

needed
Construct

Maintain

Develop

not needed Demolish

Figure 1. The life cycle of a road asset

An example to help the interpretation of Figure 2 is as follows. At t = 0, a road link consisting of a bridge
and a road section starts to provide a level of service that is higher than the initially agreed upon required
level of service (e.g. the travel time costs required per vehicle to go from A to B should be below x, and
earthquake risks should be below y). To ensure this, the road is to have a roughness above 20 mm/m,
and the bridge is designed to withstand an earthquake of magnitude 4 with a very low probability of
damage. Over time both assets deteriorate and the provided level of service decreases, i.e. due to the
worsening road condition vehicles on average take more time to go from A to B, and the earthquake
risks increase because the probability of the bridge withstanding an earthquake of magnitude 4 with no
damage decrease.

6 7
Provided
LOS
3
Implied
required
LOS 3 5
2

1 4
LOS

2
1

Intervention

Agreed Agreed required

required level
LOS

0 Time t

Figure 2. Illustration of changes in required and provided level of service

the word asset in each of the three definitions could be replaced by either the word “network”, the
word “object”, the word “element” or the word “segment”.

2
 SIAM 2017

1.3 Road asset management

Road asset management is the process used to ensure that existing road assets provide adequate
levels of service for specified periods of time. Good road asset management requires taking into
consideration the benefits and costs of the road assets to all members of society, including economic,
ecological and social benefits and costs, as well as, balancing the need for prediction accuracy with
analysis effort. Road asset management involves all phases in the life-cycle of road assets, which is
illustrated in Figure 1.
More specifically, the goal of road asset management is to determine optimal intervention programs 2,
i.e. the interventions 3 to be executed to provide stakeholders 4 with the maximum net benefit. The
implementation of these programs ensures that there is an optimal trade-off between the costs 5 of
executing interventions and the costs if no interventions are executed. The execution of interventions is
required to off-set deterioration to help ensure that it continues to provide the level of service 6 for which
it was originally designed, and to modify the infrastructure so that it can provide a new level of service.
The process of road asset management defined at this level is the same for all types of road assets and
is the same regardless of the level of automation used in the process. In order for it to be executed
effectively in today’s world, however, the use of modern technology, including the use of advanced
monitoring techniques and computer management systems are required.

2 PROCESS
The road asset management process is shown in Figure 3. The tasks are each explained in the following
sections. The speed and frequency with which this process is run varies from organisation to
organisation.

Figure 3. The road asset management process

2 Intervention program – A list of interventions to execute over a defined period of time.

3 Intervention – A group of activities that are to be executed at the same time on infrastructure.
4 Stakeholder – A group of people affected by the management of infrastructure, e.g. owner, the user

and the general public.

5 Costs – The impacts on stakeholders. They should be expressed in quantifiable units, e.g. monetary

units (m.u.). They include costs that are normally expressed in m.u. such as material costs, and costs
that are not normally expressed in m.u. such as noise costs. The costs to be considered are to be
established in a cost hierarchy. Although one can see negative impacts on stakeholders as costs and
positive impacts on stakeholders as benefits, no such distinction is made in this paper, as it is not
required from a mathematical sense, as a negative cost is a benefit.
6 Level of service – How stakeholders are affected, e.g. amount of travel time expected per unit time.

3
 SIAM 2017

3 ESTABLISH SERVICE LEVEL GOALS AND CONSTRAINTS

3.1 General
The establish service level goals 7 and constraints 8 sub process is used to determine what should be
maximised (e.g. number of vehicles per hour) or minimised (e.g. number of accidents per year) and
what has to be done (e.g. stay within the budget), when determining the intervention program to be
followed. It is an iterative process that brings together the requirements and wishes from multiple
sources and coalesces them into a common set of goals and constraints, and eliminates conflicts or
makes clear how trade-offs are to be made between them. In this process, the requirements, or wishes,
from all sources are considered, including those from 1) codes and laws, 2) politics, high level budgets,
and high level spatial planning, 3) requirements from operations, 4) requirements from new construction,
5) requirements from demolishment. It involves consideration of what the road asset management
organisation can offer or would like to offer, and consideration of what stakeholders of the infrastructure,
e.g. users, persons who are directly affected by the infrastructure but who are not users, and persons
who are indirectly affected by the infrastructure. The latter can require considerable interaction between
multiple parties and often includes dealing with stakeholder representatives, such as politicians or
lobbies, as well as through consideration of laws.
To ensure that all goals and constraints are considered in each instance of the process (sometimes they
may not be explicitly stated), it is useful to establish a cost hierarchy 9 in which all things that should be
measured are to be listed. These should be mapped to general wishes such as, e.g. the longest shortest
drive between any two nodes in the road network is to be under x hours, and the number of accidents
should be less than x per year. It should also be clear how the costs are to be measured in practice, i.e.
which service level indicators 10 are to be used.
To establish reasonable service level goals and constraints, it is useful to have well founded predictions
of what may happen in the future, e.g. how the ability of the infrastructure to provide the required service
level is expected to change and how the required service level is expected to change. This service level
is coupled with how much money is needed to intervene on the infrastructure to help ensure that it
provides an adequate service level, and what will happen if it is not available. The service level goals
and constraints are normally defined for specific time periods, e.g. five years.
3.2 Cost hierarchy
A cost hierarchy, which is a complete definition of the costs to be considered, has to be made to
objectively evaluate how well road assets provide the required service level when different monitoring
and intervention strategies 11 and programs 12 are evaluated. The cost hierarchy should be complete,
operational, quantifiable and non-redundant. It needs to be
- complete to ensure that the value of costs of one cost type is not improved to the detriment of
another. For example, the reduction of the cost of manual labour on a construction site which
results in an inadvertent rise in the number, and, therefore, the costs, of accidents.
- operational so that it can be used. For example, there is little need to have a cost hierarchy
which includes a breakdown of accidents by the age of the persons in the vehicles if it is
considered to be too laborious to collect such information.
- quantifiable as it is difficult to compare monitoring and intervention strategies and programs
using qualitative measures. In this case, it is better to make difficult to quantify costs, such as
noise, into something quantifiable and accept that there is a substantial variation in the possible

7 Service level goal – a goal that describe how stakeholders should be affected.
8 Service level constraint – Something that can prevent optimal service levels being provided, e.g.,
available financial resources.
9 Cost hierarchy – a list of all the cost types to be considered in the management of infrastructure.

10 Service level indicators – Measures used to evaluate the provided service level. They may be

qualitative or quantitative.
11 A monitoring strategy – a list of monitoring activities that should be performed depending on the state

of the infrastructure. This includes the collection, analysis and evaluation of the state of the infrastructure
12 A monitoring program – a list of monitoring activities to be performed in a specified time period.

4
 SIAM 2017

values than to simply say, for example, that one intervention program would result in less noise
than another. The two main possibilities to measure things in proportional units are utility or
m.u.. The latter has the advantage that many costs are already measured in terms of m.u..
- non-redundant so that accurate results in analyses are obtained.

Examples for road networks include [7], [8], [9], [10].

3.3 Service level indicators

Service levels can be defined in terms of costs directly or in terms of the values of service level
indicators. For example, one could say that on average the travel time costs of a vehicle between point
A and point B are to be less than x m.u., which is given by y persons per vehicle multiplied by z vehicle-
hours per trip multiplied by aa m.u. per person-hour spent travelling. One could, however, admit that it
would be hard to keep track of this all of the time, and say that this number was to be collected for a test
vehicle travelling from point A to point B at specified times over the course of a month, which would be
a service level indicator, i.e. it doesn’t tell you exactly the service level to be provided but it does give
you an indication of the service level being provided. One could also say that although we care about
the total amount of travel time of the vehicles, that we would use how often one can travel from point A
to point B in x time units, e.g. if we drove a test vehicle from point A to point B 10 times and 9 times out
of 10 it was below our threshold we could define this as an acceptable service level and, therefore, the
reliability of our test would be our service level indicator. Four common service level indicators for
infrastructure are reliability, availability, maintainability and safety (Table 1).
Table 1. Four common service level indicators

Term Definition Useful in appropriate situations

Reliability the probability that infrastructure when the costs related to the infrastructure
provides a specific level of service for being non-operational are substantially
a specified period of time higher than when it is operational, e.g. in
the case of sudden deterioration of
infrastructure due to floods

Availability the amount of time that infrastructure used when the costs related to the
provides a specific level of service over infrastructure being operational and non-
a specified period of time taking into operational are relatively similar, e.g. in the
consideration the length of time that case of gradual deterioration of
the infrastructure is not providing the infrastructure due to chloride induced
specific level of service due to the corrosion
execution of interventions

Maintainability a measure of the costs associated with used when the costs related to the
the execution of interventions or the execution of interventions on the
time required to execute interventions infrastructure is of particular interest
over a specified period of time

Safety determined through the assessment of used when the number of accidents is too
the number of persons injured high

A significant part of the estimation of the values of service level indicators for road networks depends
on their topology. For example, the estimation of the reliability of travelling from A to B in a road network
depends on whether A and B are connected in a series, parallel, or a more complex network, as well
as, how each of the objects in each of the network links performs. Similar considerations are applicable
to the estimation of availability and maintainability. The task of determining the service level indicators
to be used involves consideration about how information is collected, i.e. monitoring, and making
assumptions about what is important. This includes the generation and assessment of possible

5
 SIAM 2017

monitoring strategies, the estimation of the costs associated with the performing monitoring activities,
the development of data management strategies, and consideration as to how the data will be used.
3.4 Changes over time
The service level to be provided by infrastructure changes over time. It may change due to changes in
already defined cost types, e.g. last year travel time costs were not to exceed x m.u. and this year they
are not to exceed y m.u.. It may also change due to changes in not yet defined requirements, e.g. last
year there were no limitations on the amount of CO 2 emitted, and this year the amount of CO 2 emitted
is not to exceed x tons.
Changes in the service level to be provided can be classified as gradual or sudden. A gradual change
is one that happens in a way that there is enough time to execute an intervention so as to ensure that
the infrastructure continues to provide an adequate service level. A sudden change in the service level
is one that happens in a way that there is not enough time to execute an intervention so as to ensure
that the infrastructure continues to provide an adequate service level. For example, a gradual change
may be the increases in the amount of travel time due to increasingly congested roads, something that
may happen at a rate of 2% per year. A sudden change may be a new law that allows the axle loads to
increase by 20%. The classification of a change in service level as gradual or sudden depends on the
state of preparedness and the perception of the road asset management organisation.
3.5 Conclusion
There is little work in this direction in terms of automation, but the establishment of the cost hierarchy,
and associated indicators, to be used throughout the road management organisation to make decisions,
and to communicate to both internal and external stakeholders is essential in the facilitation of the
automation of road asset management, as well as simply good road asset management.

4 ESTABLISH STRUCTURES, PROCESSES, MODELS AND STRATEGIES

4.1 General
The establish structures, processes, models and strategies sub process sets the situation in which the
monitoring and intervention programs are to be constructed (Table 2).

6
 SIAM 2017

Table 2. Structures, processes and models

Item Covers Includes
Structures the the structure of
aspects of - the infrastructure management organisation itself,
the - the information to be used (sometimes referred to as an
situation
infrastructure model),
that are
relatively - the things that are to be considered in determining the optimal
static monitoring and intervention programs, i.e. the cost hierarchy.

Processes the the processes to be used to:

aspects of - predict the values of service level indicators,
the - determine the monitoring strategies and programs, e.g. [15],
situation
- determine the intervention strategies and programs,
that are
relatively - analyse the effectiveness of monitoring activities and
dynamic interventions, and
- analyse the effectiveness of the road asset management
process.

Models the principally the models to be used to help

models to - make predictions of what will happen in the future, e.g. the future
be used required service level, the future ability of the infrastructure
in the
provide the service level, and the costs associated with states of
processes
the items, and
- determine optimal strategies and programs, e.g. optimal
monitoring strategies, optimal intervention strategies, optimal
monitoring programs, and optimal intervention programs

It is convenient to group the prediction models with respect to the speed of the processes they are being
used to model, i.e. gradual and sudden. A gradual process is one that happens in a way that there is
enough time to execute an intervention so as to ensure that the infrastructure continues to provide an
adequate level of service. A sudden process is one that happens in a way that there is not enough time
to execute an intervention so as to ensure that the infrastructure continues to provide an adequate level
of service. An example of a gradual process is chloride induced corrosion of reinforced concrete in which
an expansive rust product is created. An example of a sudden process is an earthquake.
Popular models for gradual processes include analytical models, regression models, and Markov
models. Analytical models are used for example to predict the speed of chloride ingress into reinforced
concrete bridge decks, the start of corrosion, and the speed with which the cross section area of the
reinforcement will be reduced. Popular models for sudden processes include event trees, e.g. [11], fault
trees, e.g. [12], and Bayesian network models, e.g. [13].
Popular models to be used to determine optimal strategies and programs can be grouped as standard
mathematical models, e.g. block replacement [14], or operations research models, e.g. specific
mathematical models used to solve specific problems.
The word “strategies” covers the determination of both the monitoring strategies 13 and intervention
strategies 14 to be followed for each type of asset. These are explained in more detail in the following
section.

13 A list of monitoring activities that should be performed depending on the state of the infrastructure.

This includes the collection, analysis and evaluation of the state of the infrastructure.
14 A list of interventions that should be executed on infrastructure depending on the state of the

infrastructure.

7
 SIAM 2017

4.2 Strategies
Monitoring strategies include all monitoring activities to be performed on road assets taking into
consideration many possible states of the asset. These are established independently of their actual
state. When they are developed, consideration is to be given to the range of possible monitoring
activities, the effectiveness of the monitoring activities in each possible state and the associated costs.
Monitoring strategies are developed without consideration of externalities, such as the state of other
road assets or assets in other infrastructure networks.
The task to develop and evaluate monitoring strategies involves consideration of the information that
will be collected and how this information will be used to affect the development of intervention
programs. It also involves the estimation of the costs associated with performing the monitoring
activities, their effects on service levels, their duration, and the probability of them obtaining information
as intended, as well as the consequences if they do not work as intended.
Intervention strategies include all interventions to be executed on an asset taking into consideration
many possible states of the asset or many possible situations in which the asset could be. For example,
if there is cracking in the pavement part of a road section, a crack filling intervention should be executed
and if there is cracking and significant unevenness then a partial depth repair intervention should be
executed. Intervention strategies are developed without consideration of externalities, such as the state
of other road assets or other infrastructure networks.
The task to develop and evaluate intervention strategies involves consideration of how the interventions
affect the present and the future. For example, if one is concerned with loading carrying capacity, the
addition of a new concrete layer on a reinforced concrete element, may be seen as having an effect on
both the present and future behavior of the element, and the addition of a waterproofing layer on a
concrete bridge deck, may be seen as having an effect on only the future. It also involves the estimation
of the costs associated with the execution of the interventions, their effects on future levels of service,
their duration, and the probability of them working as intended, as well as the consequences if they do
not work as intended.
The generation of possible intervention strategies requires in-depth asset specific knowledge. The
generation of such strategies requires good technical persons and organisations with structures that
encourage both cooperation and innovation.
The evaluation of the possible strategies and the determination of the optimal ones require a good
foundation in decision theory and an ability to model how costs will change over time if each of the
strategies are followed. Some models used to evaluate strategies are common, such as block
replacement models (e.g. [16]), age replacement models (e.g. [17]), and Markov models (e.g. [18]).
Others are custom built to address specific questions, using standard operations research methods (e.g.
[19]). In the evaluation of strategies both the costs during execution of interventions and between the
executions of interventions must be considered, e.g. [20].
If strategies are focused on a particular part of the life cycle of a road asset that can be referred to using
the name of the activity to which they are associated as an adjective, e.g. a maintenance intervention
strategy 15 or a development intervention strategy 16.
4.3 Conclusion
Once the structures, processes, models and strategies are determined, they are regularly checked to
ensure that they are still valid and only updated when necessary. They are, in general, updated at much
lower frequency than the frequency of the determination of, for example, intervention programs.

15 A maintenance intervention strategy is one of how to modify existing infrastructure so that it continues
to provide the same level of service as initially intended. It principally deals with possible changes in the
infrastructure. For example, if more than 50% of the material from a track is worn away then the track
will be replaced.
16 A development intervention strategy is one of how to modify existing infrastructure so that it provides

a new level of service. It principally deals with possible changes in requirements. For example, if the
number of trains travelling on a railway line is over 90% of the design capacity then a new track will be
built.

8
 SIAM 2017

This sub process ends with the fixed structures, processes, models and strategies to be used to develop
the monitoring and intervention programs to be followed.

5 CONSTRUCT MONITORING PROGRAMS

The construct monitoring programs sub process is an iterative process that includes:
- the identification of the assets on which a monitoring activity is to be performed taking into
consideration the actual state of the asset and the strategy to be followed,
- the construction of candidate monitoring programs, which can be done bottom-up or top-down,
where bottom-up refers to identifying all assets that are to have a monitoring activity and then
combining them in a plan to perform them within a specific time period, and top-down refers to
identifying the general areas to have monitoring activities performed in a specific time frame
and then selecting the specific monitoring activities to perform.
- the verification of the feasibility of the candidate monitoring programs, e.g. are enough persons
and equipment available, are there possible synergies with other types of activities that would
make it better to perform the activities in a different order?
- the determination of the optimal monitoring program.

The results of this sub process is implementable monitoring programs. Some examples of work done to
wards the automation of the determination of optimal monitoring programs include [21], [22], [23].

6 PERFORM MONITORING ACTIVITIES

The perform monitoring activities sub process includes all activities following the development of the
monitoring program to perform the monitoring activity and to verify that the monitoring activity was
performed satisfactorily. To help ensure successful automation of this sub process the expectations
from the monitoring activities need to be clearly defined, e.g. the costs and the accuracy of information.
Some work towards the automation of activities include that being done in the areas of robotics, e.g.
[24], [25].

7 CONSTRUCT INTERVENTION PROGRAMS

The construct intervention programs sub process includes the following activities:
- The identification of the assets that are to have an intervention according to the current state of
the assets and the intervention strategies.
- The construction of candidate intervention programs, which can be done either bottom-up or
top-down, where bottom-up refers to constructing the programs thinking first as to what could
be done on the asset level, and top down refers to constructing the programs thinking first as to
what could be done on the network level.
- The verification of the feasibility of the candidate intervention programs, e.g. are enough people
and equipment available, are synergies possible?
- The determination of the optimal intervention program, i.e. the one that provide the optimal
service level.

Intervention programs include the interventions to be executed on assets with consideration of

externalities, such as the states of other assets within the same network or other assets in other
networks. The consideration of externalities results in deviations from those that would result if optimal
intervention strategies were followed without consideration of externalities. For example, if an asset is
not in a state for which the intervention strategy includes the execution of intervention, but part of the
network is closed due to the execution of other interventions, than an intervention might also be
executed on the asset.
The development of optimal intervention programs is dependent on the location of the assets in networks
and the timing of the interventions on the items. The determination of optimal intervention programs
requires making trade-offs and deviations from the theoretically optimal intervention strategies.

9
 SIAM 2017

The generation of possible intervention programs requires good technical persons and organisations
with structures that encourage both cooperation and innovation. One of the particular challenges in the
generation of such programs that give adequate consideration to interventions, is that it requires
understanding of:
- how the asset needs to be adapted or could be adapted to ensure that it provides adequate
levels of service,
- the costs related to the execution of interventions and how they change over time,
- how the costs related to the execution of interventions are affected by the execution of other
interventions simultaneously.

The generation of possible programs increases in difficulty with the size of the road asset management
organizations, as there is an increasing division of labour, i.e. different persons are responsible for the
execution of maintenance interventions 17 and development interventions 18, and different persons are
responsible for different types of assets. This difficulty can, however, be overcome with clear guidelines
of how information is to be aggregated and how decisions are to be made as to which interventions are
to be included in the programs.
Traditionally intervention programs have been developed through negotiations with these different
persons. Increasingly, however, computers are used. Over time the algorithms used are increasingly
more sophisticated. Some example include [26], [27], [17].
This sub process, also, involves securing the financing to implement the programs, ensuring that all
legal requirements are met and ensuring that there is enough staff available for implementing the
program. Increasing numbers of computer tools are being used to facilitate this.

8 PLAN AND EXECUTE INTERVENTIONS

The plan and execute interventions sub process includes all activities from completion of the intervention
programs to supervision of the execution of the interventions. This sub process involves tendering the
interventions, reviewing the applications, selecting the best companies to execute the interventions,
planning the interventions, communicating the activities with stakeholders, and all other normal project
management related activities. Some examples towards the automation of this portion of the process
include [28], [29], [30].

9 ANALYSE EFFECTIVENESS
The analyse effectiveness sub process includes the analysis of the effectiveness of monitoring activities
and interventions. The effectiveness of monitoring activities is analysed to compare the actual completed
monitoring activities with the expected monitoring activities, with respect to their costs and the accuracy
of their results. The effectiveness of interventions is analysed to compare the actual completed
interventions with the expected interventions, with respect to their costs and their effect on the
infrastructure. The results of this task can be used to modify existing monitoring programs, but can also
be used to update the information used in the processes to establish service level goals and constraints,
processes, models and strategies. An essential component to help automate this sub process is the
definition of how effectiveness is to be measured and reported, and how the information is to be used.
An example of work heading in this direction is [31].

10 ANALYSE PROCESS
In the analyse process sub process, all of the activities of the process and how the process as a whole
works is checked to ensure that it is working as expected and to see if possible improvement can be
made. When deviations are found either the process should be modified or the process should be
encouraged to execute the process as intended. When possible improvements are found the process

17 Maintenance intervention – An intervention that does not alter the original functionality of the item.
18 Development intervention – An intervention that does alter the original functionality of the item.

10
 SIAM 2017

should be modified, if the benefits of modification outweigh the costs. Some examples of work heading
in this direction are [32], [33], [34].

11 CONCLUSION
Road asset management is an important process, but one that is requires substantial human effort and
is open to human interpretation. Over time, this process will become increasingly automated, allowing
humans to maximise their net benefit from their infrastructure, and simultaneously reducing the amount
of human effort required to do so. The complete high level process presented in this paper is to be used
to enable this automation of road asset management. It should be used by road asset managers who
are establishing their own more detailed process to ensure that all involved stakeholders agree as to
the activities involved and the reasons why.

12 REFERENCES

1. OECD. (2001), Asset Management for the Roads Sector, OECD Publications Service,
Paris, France, 81 pages.
2. ISO 55001, (2014), Asset management – Management systems – Requirements, ISO
copyright office, Geneva, Switzerland, 22 pages.
3. UK Roads Liason Group, (2013), Highway Infrastructure Asset Management: Guidance
Document, Queen’s Printer and Controller of Her Majesty’s Stationery Office, London,
United Kingdom, 136 pages.
4. Adey B.T., Klatter, L., Kong, J., (2010), The IABMAS Bridge Management Committee
Overview of Existing Bridge Management Systems, 99 pages.
5. Santos, J., Ferreira, A., Flintsch, G., (2015), life cycle assessment model for pavement
management: road pavement construction and management in Portugal, International
Journal of Pavement Engineering, 16, p. 315-336.
6. Xu, H., Pan, Y., Wang, Q., Qian, Y., Dong, P., (2013), Expressway Electro-Mechanical
Equipment Maintenance Management System Functional Research Based on B/S, 4th
international conference on transportation engineering, Chengdu, China October 19-20
7. Adey, B.T., Lethanh, N., Lepert, P., (2012), An impact hierarchy for the evaluation of
intervention strategies for public roads, European Pavement Asset Management
Conference, Malmö, Sweden, September 5-7.
8. Adey, B.T., Herrmann, T., Tsafatinos, K., Luking, J., Schindele, N., Hajdin, R., (2010).
Methodology and base cost models to determine the total benefits of preservation,
interventions on road sections in Switzerland. Structure and Infrastructure Engineering:
Maintenance, Management, Life-Cycle Design and Performance 8, 639-654.
9. ECOPLAN, (2010). Sustainability indicators for road infrastructure projects, in: ECOPLAN,
(Ed.), Handbuch eNISTRA. ECOPLAN, Bern, Switzerland.
10. New Zealand Transport Agency, (2010). Economic evaluation manual. New Zealand
Transport Agency.
11. Adey, B.T., Hajdin, R., Birdsall, J., (2009), Methodology to determine optimal intervention
strategies for structures adversely affected by latent processes, Transportation Association
of Canada, Annual Conference, Vancouver, Canada, October 19-22.
12. Johnson, P., (1999), Fault tree analysis of bridge failure due to scour and channel instability,
Journal of Infrastructure Systems, Vol. 5, No. 1, March.
13. Deublein, M., Schubert, M., Adey, B.T., Garcia de Soto, B., (2015), A Bayesian Network
Model to Predict Accidents on Swiss Highways, Infrastructure Asset Management. 2(4), pp.
145–158.
14. Adey, B.T., Lethanh, N., Kielhauser, C., (2014), Comparison of three preventive models to
determine optimal intervention strategies for transportation infrastructures in alpine regions,
Architecture, Engineering and Construction, 3:3, September, 153-167.
15. Adey, B.T., Hackl, J., Lam, J.C., van Gelder, P., van Erp, N., Prak, P., Heitzler, M., Iosifescu,
I., Hurni, L., (2016), Ensuring acceptable levels of infrastructure related risks due to natural
hazards with emphasis on stress tests, 1st International Symposium on Infrastructure Asset
Management (SIAM), Kyoto, Japan, January 21-22.

11
 SIAM 2017

16. Adey, B.T., Lethanh, N., Kielhauser, C., (2014), A comparison of the suitability of age
replacement, block replacement and Markov models in the determination of optimal
intervention strategies for road infrastructure in alpine regions, Brenner Congress 2014,
Innsbruck, Austria, Feb 20-21.
17. Adey, B.T., Hajdin, R., (2011), Methodology for determination of financial needs of
gradually deteriorating bridges with only structure level data, Structure and Infrastructure
Engineering, 7:8, 645–660.
18. Lethanh, N., Adey, B.T., Fernando, D., (2015), A Markov chain model to determine
optimal intervention strategies for multiple objects affected by uncorrelated manifest and
latent processes, Structure and Infrastructure Engineering, 11:3, March, 389-401.
19. Mirzaei, Z., Adey, B.T., (2015), Investigation of the use of three existing methodologies to
determine optimal life cycle activity profiles for bridges, Structure and Infrastructure
Engineering, 11:11, November, 1484-1509.
20. Adey, B.T., Lethanh, N., Hartmann, A., Viti, F., (2014), Evaluation of intervention
strategies for a road link in the Netherlands, Built Environment Project and Asset
Management, 4:2, 180 – 198.
21. Jamali, A., Adey, B.T., (2016), The effect of variations in management decision processes
on the management of bridges, Civil Engineering and Management, 22:1, 92-104.
22. Konakli, K., Sudret, B., Faber, M.H., (2015), Numerical Investigations into the Value of
Information in Lifecycle Analysis of Structural Systems Asce-asme Journal of Risk and
Uncertainty in Engineering Systems, Part A: Civil Engineering, Volume 2, Issue 3.
23. Lethanh, N., Richmond, C., Adey, B.T., (2016), Investigation of the ability to estimate
values of road section condition indicators based on their spatial correlation, Journal of
Infrastructure Systems, 22:3, 04016006.
24. Kielhauser, C., Romer, F., Adey, B.T., (2017), Preliminary estimates of the viability of
UAV-based bridge inspections in Switzerland, Fourth Conference on Smart Monitoring,
Assessment and Rehabilitation of Civil Structures, SMAR 2017, Zurich /Switzerland, 13 –
15 September.
25. Lattanzi, D., Miller, G., (2017), Review of Robotic Infrastructure Inspection Systems,
Journal of Infrastructure Systems, Volume 23, Issue 3.
26. Lethanh, N., Adey, B.T., Burkhalter, M., (2016), Determining optimal work-zones on large
infrastructure networks in a geographic information system, International Symposium on
Infrastructure Asset Management (SIAM), Kyoto, Japan, January, 21-22.
27. Lethanh, N., Adey, B.T., Sigrist, M., (2014), A mixed-integer linear model for optimizing
work zone interventions on a transportation network, International Conference on
Engineering and Applied Sciences Optimization, Kos Island, Greece, June 4-6.
28. Bosche, F., Teizer, J., Haas, C.T., Caldas, C., (2006), Integrating data from 3D CAD and
3D Cameras for real-time modelling, Proceedings of joint international conference on
computing and decision making in civil and building engineering, Montréal, Canada, June
14-16.
29. Jin, X.H., Le, Y., (2014), Monitoring construction projects using information technologies,
Proceedings of the 17th International Symposium on Advancement of Construction
Management and Real Estate, p. 1011-1020.
30. Aliverdi, R., Naeni, L.M., Salehipour, A., (2013), Monitoring project duration and cost in a
construction project by applying statistical quality control charts, International Journal of
Project Management, Volume 31, Issue 3, p. 411-423.
31. Li, J., Luhr, D. R., Uhlmeyer, J.S., Mahoney, J. P., (2013), Evaluation of maintenance
effectiveness for WSDOT pavement network, Transportation Research Record,
32. Richmond, C., Kielhauser, C., Adey, B.T., (2016), Developing performance measures for
road network managers facing diverse environments, Benchmarking an international
journal, 23:7, 1876-1891.
33. Richmond, C., Kielhauser, C., Adey, B.T., (2015), Developing key performance indicators
for cantonal road management in Switzerland, The Asset Management Conference,
London, United Kingdom, November 25-26.
34. Richmond, C., Kielhauser, C., Adey, B.T., (2014), Adding realism to cost functions in
sustaining long lasting infrastructures to identify better management, Industry Studies
Association, Portland, Oregon, May 27 – 30.

12