Roads Network Management with System Dynamics: Where Detours and U-turns are Mandatory


Introduction

"Life can be so nonlinear", says Kovach (1960) in his famous essay with the message that nonlinear problems are prevalent in various fields and require new approaches and tools for their solution. The author argues that traditional linear methods are inadequate for dealing with nonlinear problems and that a deeper understanding of nonlinear phenomena is necessary for progress in science and technology. 

The transportation problems are rooted in the basic structure of the system. The solution provided to a problem may create another problem elsewhere (Abbas & Bell, 1994). The traditional linear approach undermines many soft variables that are key to the system's behaviour; therefore, linear models do not precisely represent the reality of the road transport system. On the other hand, system dynamics is designed to analyze complex systems governed by nonlinear behaviours and feedback mechanisms, making it a more appropriate methodology for transportation modelling.

This essay explores the commonly understood linear roads and their management in a nonlinear, chaotic world. It explores various aspects of road network management problems and presents arguments requiring a deeper understanding of nonlinearity before jumping into management solutions.


A Few Key Terms

Road Networks are the networks of various roads based on their functions within geography. These networks are complex and usually large-scale systems. These systems are affected by other modes of transport, such as waterways and airways. The management of road networks involves multiple stakeholders, including the owners at different levels of the government and road users such as drivers, passengers and pedestrians. The demand and supply of the road network depend on various sectors of the economy, which itself is a complex phenomenon. 

The system is a bounded repetitive pattern of behaviour that can be modelled. The parts or structures interact in the system to advance the system's state and produce outcomes (Berrisford, 2022). The system approach helps represent and simulate complex systems through effective and efficient processes and further assists in making policy decisions (Abbas & Bell, 1994).

System Thinking and System Dynamics – System thinking involves thinking, talking and observing systems for decision-making. Whereas system dynamics include quantitative aspects of system analysis, help design policies and evaluate those policies under different circumstances through simulation. According to Donella Meadows, system dynamics modelling provides an understanding of the system through modelling, analysis of the policies and addressing implementation issues. 

Linearity and Nonlinearity – The linear method of observing and addressing issues involves simplifying the problem and assuming the static process, which undermines the dynamic behaviour of the process and neglects relevant variables. Scientists and practitioners tend to ignore the effect of variables that may sway away from their linear model. Linear thinking assumes that there is a problem and a cause for the problem. The linear method tends to offer closed-form solutions that neglect the system's dynamic behaviour, where the offered solution may further create new problems or reinforce the existing problem in the long run. Linear thinking is uni-directional and does not consider the existence of the feedback. On the other hand, nonlinear thinking considers the effect of the system's dynamic behaviour, including continuous feedback that makes the system vibrant. Nonlinear methods are closer to reality and usually as complex as the truth is.


Why Linear Methods may not be Appropriate for Roads Network Management? 

Linear methods may not be able to account for the nonlinear relationships between transport variables as transport systems are complex and dynamic. Transportation systems are subject to a wide range of factors that can affect their behaviour, such as changes in traffic volume and patterns, engineering, road users behavior, weather conditions, climate change & and disasters, other modes of transport, socio-economic conditions and many more. Linear models assume that the relationships between these variables are constant over time, which is not always the case as they are adaptive, counterintuitive, and self-corrective. Due to massive and complicated management challenges, transportation modelling becomes difficult. One of the major issues with linear methods is that they tend to oversimplify complex systems and ignore critical qualitative dimensions. It also does not account for nonlinear feedback loops, which are common in the transport system and lead to inaccurate predictions and decision-making. Linear methods cannot capture the dynamic behaviour of the system and the interdependencies between its components. The input of road users to the decision process (such as renewals or maintenance decisions) operates within a much fuzzier and qualitative political environment (Abbas & Bell, 1994; Guevara et al., 2017; Linard, 2000, 2009; Ruiz & Guevara, 2020; Shiboub & Assaf, 2022).


Why System Dynamics (SD) for Roads Network Management?


SD helps understand and control real-world complex and dynamic systems 

System dynamics is designed to help understand the dynamics of different real-world systems. It is a methodology through which powerful management tools are developed to enhance the ability to control complex systems. SD furnishes a logical, systematic and detailed representation of complex, large-scale systems. The transportation system is a perfect example of such a dynamic and complex system (Abbas & Bell, 1994).

A linear model would not be able to capture the system's dynamic behaviour. The System Dynamics approach is better suited for modelling complex systems with feedback loops and nonlinear relationships. The System Dynamics model can capture the interactions between different components of the transport system, such as the physical condition of the roads, the level of service provided, and the maintenance costs (Bjornsson et al., 2000). 

Pavement maintenance is an example of a more extensive complex road network component. It is part of a complex system with significant feedback and various soft variables. One standard soft variable is the political environment upon which the pavements need to be maintained (Linard, 2000). Pavement maintenance is a low-priority area of government funding where the response time is usually lengthy. System dynamics help understand the phenomenon by focusing on crucial stocks and flows (Linard, 2009).

The approach assumes that the systems are formed by a series of components that are continuously evolving over time as a result of causal interdependencies and feedback relationships between system components (Guevara et al., 2017; Ruiz & Guevara, 2020)

Pavement maintenance can be an example of a complex road transport system with significant feedback, making it a suitable field for system dynamics enquiry. The system dynamics-based pavement management model can help prioritize rehabilitation treatments based on user preferences and budget constraints and identify the consequences of different budgetary approaches. The model can provide feedback to decision-makers, including the number of households served by very rough roads, the number of user complaints, roughness-related accidents and vehicle operating costs (Linard, 2009).

SD helps design, formulate and test different scenarios and policies

System dynamics has a vital role in policy analysis. It can be used to design, formulate, and test different scenarios and policies. It provides valuable information to policy- and decision-makers, thus supporting the decision-making process in strategic planning (Abbas & Bell, 1994). These models can help evaluate the impacts of different policies and interventions on transportation systems and identify potential unintended consequences.

System dynamics models have been used to evaluate the impacts of different policies and interventions on transportation systems, such as road pricing, public transportation subsidies, and fleet maintenance policies. System dynamics helps explore the real drivers of future demand and explain how to change user perception and behaviour (Shepherd, 2014).

As per one of the studies, the system dynamics model is used for performing an integrated analysis of sustainable pavement development, considering construction and maintenance activities, CO2 emissions, and costs associated with the development of roadways. The model is used to evaluate pavement networks' environmental and economic impact and identify the most effective interventions for reducing emissions and costs (Ruiz & Guevara, 2020).

As per another similar study,  the System Dynamic Model is used for Sustainable Road Rehabilitation by integrating technical, economic, and environmental factors to identify the best road strategies for network-level maintenance and rehabilitation. The model was tested and validated, where it identified optimal pavement solutions and pavement management systems policies for improved sustainability. The model recommended three optimal pavement solutions: asphalt with a 10-year rehabilitation cycle, concrete with a 20-year rehabilitation cycle, and a combination of asphalt and concrete with a 15-year rehabilitation cycle. The model recommended three pavement management systems policies: determining the optimal budget allocation and average time to rehabilitation, promoting maintenance over rehabilitation, and combining budget optimization and maintenance. The author recommended that the model be used to analyze and improve road systems in other regions and contexts and help decision-makers balance technical, economic, and environmental considerations for sustainable road rehabilitation (Shiboub & Assaf, 2022).

As per another study in the U.S. highway system,  they have found that short-term reactive efforts lead to long-term deterioration of the system. The study shows that focusing on fixing major problems/failures before anything else is not an effective strategy for maintaining the highway system. The results suggest that preventive maintenance is a more effective strategy for maintaining the highway system in good condition and reducing future rehabilitation expenditures. The study proposed policies to incentivize preventive maintenance practices and reduce the reliance on significant rehabilitation efforts. The simulation model developed in the study can also be used to evaluate the effectiveness of different policies and investment strategies for maintaining the highway system (Guevara et al., 2017).

One of the research presented a paper on SD and road transport that can be used as a decision support system to allocate road maintenance funds effectively. The focus of the study was on the distribution of budget, particularly between different pavement maintenance activities. The model has been tested for several different maintenance policies, and the system's behaviour was studied for several budget allocation combinations used in the simulation experiments. The aim of the research was not to solve a single problem or optimize the maintenance budget for a particular year; instead, the main objective was to provide a decision tool that can be used to answer several what-ifs. As suggested by the author, the model can be used to identify the most advantageous rehabilitation and maintenance strategy, given the budget and other constraints (Bjornsson et al., 2000).

SD helps to look at transport management problems in a holistic approach

System dynamics takes a holistic approach to transportation planning and policy-making, which involves considering the interdependencies and trade-offs between different aspects of the transportation system. A whole system approach to transportation planning and policy-making is crucial for achieving sustainable and efficient transportation systems (Shepherd, 2014).

Using a holistic approach, system dynamics models have successfully been applied in various areas of transportation, including traffic flow, public transportation, logistics, and safety. Transportation systems involve multiple stakeholders and feedback loops with different time lags. It can help in understanding the dynamic tendencies of transportation systems and identifying the parameters that play a significant role in the stability and response of the system (Shepherd, 2014).

Not only that, system dynamics has the capability of analyzing massive and complicated management challenges. It helps increase model validity and effectiveness of decision-making processes (Shiboub & Assaf, 2022). The so-called holistic approach involves a modelling that captures several causal relationships between variables associated with highway deterioration, road funding mechanisms, highway aging, and maintenance expenditures (Guevara et al., 2017). 

SD helps to capture, visualize and simulate models and helps understand how system 
changes and predict the long-term behaviour over time of its components 

Jay Forrester, the father of system dynamics, once said, “If you can’t explain it to your barber, maybe you should be careful.” As per him, one of the aspects of SD is that though it represents complex systems, it pursues simplicity. He suggests to use the parsimony principle and asks if the model can be explained to a high school student or a layman. If not, one shall pause there and rethink.  

System Dynamics is a simulation-based technique well-suited to understanding strategies' dynamics and their effects over time. System Dynamics can capture, visualize, and simulate the causal relationships that influence the performance of interest. System Dynamics can provide insights into the long-term behaviour of a system, which is vital for asset planning (Elsawah et al., 2019).

The methodology uses simulation models as a means to examine dynamic problems. Simulation results allow the highlighting of counterintuitive behaviours and offer novel insights into the underlying structures of social systems (Guevara et al., 2017). The models are presented in graphical stocks and flow charts. The graphical interface makes the relationships between key variables apparent to the decision-makers (Linard, 2009).

System dynamics models can be built with stakeholders' input and used in the form of, for example, games or flight simulators. This simulation helps learn different policy choices and their consequences when implemented. System dynamics models can be used hierarchically to allow systems and policies to interact across space and time, which is well-suited to transportation problems (Shepherd, 2014). For example, the System Dynamics model can simulate different road maintenance policies and budget allocations and evaluate their long-term effects on the road transport system (Bjornsson et al., 2000).

SD helps consider both controllable and non-controllable factors

System dynamics can address both the 'hard' quantitative and soft' qualitative dimensions of the system. "Soft" (qualitative) data, essential in decision-making, can be readily incorporated into the model. System dynamics is instrumental in understanding the linkages between the qualitative and the quantitative aspects of road asset management (Linard, 2000; Linard, 2009). In other words, as some scholars have coined, SD incorporates controllable and uncontrollable factors (Shiboub & Assaf, 2022).


What are the Limitations?

However, there are some limitations to using SD for transportation modeling, such as its focus on the time dimension and its difficulty in accounting for spatial aspects and distribution effects. To overcome these limitations, researchers should explore and develop simple, transparent, creative, and user-oriented techniques for modelling transportation systems (Abbas & Bell, 1994).

Some challenges and limitations associated with using system dynamics models in transportation include data availability, model validation, and stakeholder engagement. There are a few recommendations for future research in this area, such as developing more integrated models that capture the interactions between different modes of transportation and incorporating uncertainty and risk analysis into the models (Shepherd, 2014).


Conclusion 

System dynamics is useful for modelling and analyzing transportation systems because it captures the complex, dynamic, and nonlinear relationships between different system elements. SD can help identify controls in the transport system being modelled, which can guide policy-makers in managing and controlling different parameters and structures of the system to improve its performance.

It can provide a foundation for structuring thoughts and better understanding complex transportation systems and their underlying problems. Comprehensive analytical models are required to simulate the nature and dynamics of the decision-making process and to develop transport management tools to ease the dilemma faced by transportation decision-makers when attempting to reach rational, informed decisions. There is a genuine need to develop transport models that show the consequences of a broad range of transport investment policies.

In conclusion, System dynamics modelling can provide a more comprehensive and integrated approach to road transport management and is a promising approach for road transportation modelling. However, it should be used with other methodologies and techniques to provide a more comprehensive and accurate understanding of transportation systems.


References 

Abbas, K. A., & Bell, M. G. H. (1994). System Dynamics Applicability to Transportation Modeling. In Transpn. Res.-A (Vol. 28, Issue 5).

Berrisford, G. (2022). 150 systems thinking terms | LinkedIn. https://www.linkedin.com/pulse/systems-thinking-terms-concepts-graham-berrisford/?trackingId=7glA6gtjTRQebRLt%2FFWQVw%3D%3D

Bjornsson, H. C., De La Garza, J. M., & Nasir, M. J. (2000). A Decision Support System for Road Maintenance Budget Allocation.

Elsawah, S., Danesh, D., & Ryan, M. (2019). A strategic asset planning decision analysis: An integrated System Dynamics and multi-criteria decision-making method.

Forrester, J. W. (2009). Some Basic Concepts in System Dynamics. Sloan School of Management, MIT.

Guevara, J., Garvin, M. J., & Ghaffarzadegan, N. (2017). Capability Trap of the U.S. Highway System: Policy and Management Implications. Journal of Management in Engineering, 33(4). https://doi.org/10.1061/(asce)me.1943-5479.0000512

Kovach, L. D. (1960). Life can be so Nonlinear. Sigma Xi, The Scientific Research Society.

Linard, K. T. (2000). Application of System Dynamics to Pavement Maintenance Optimisation.

Linard, K. T. (2009). Application of System Dynamics to Unsealed Pavement Maintenance. In International System Dynamics conference Albuquerque.

Ruiz, A., & Guevara, J. (2020). Environmental and Economic Impacts of Road Infrastructure Development: Dynamic Considerations and Policies. Journal of Management in Engineering, 36(3). https://doi.org/10.1061/(asce)me.1943-5479.0000755

Shepherd, S. P. (2014). A review of system dynamics models applied in transportation. Transportmetrica B: Transport Dynamics, 2(2), 83–105. https://doi.org/10.1080/21680566.2014.916236

Shiboub, I., & Assaf, G. J. (2022). System Dynamic Model for Sustainable Road Rehabilitation Integrating Technical, Economic, and Environmental Considerations. Journal of Management in Engineering, 38(5). https://doi.org/10.1061/(asce)me.1943-5479.0001060

 


Comments