Samer Madanat

Optimal Inspection and Repair Policies for Infrastructure Facilities

State-of-the-art decision-making models in the area of infrastructure maintenance and rehabilitation (which are based on the Markov Decision Process) do not take into account the uncertainty in the measurement of facility condition. This paper presents a methodology, the Latent Markov Decision Process (LMDP), which explicitly recognizes the presence of random errors in the measurement of the condition of infrastructure facilities. Two versions of the LMDP are presented. In the first version, the inspection schedule is fixed, which is the usual assumption made in state-of-the-art models. The second version of the LMDP minimizes the sum of inspection and M & R costs. An empirical comparison of the two versions of the LMDP and the traditional MDP illustrates the importance of incorporating measurement uncertainty in decision-making and of optimizing the inspection schedule.

Perception updating and day-to-day travel choice dynamics in traffic networks with information provision

A Bayesian updating model is developed to capture the mechanism by which travelers update their travel time perceptions from one day to the next in light of information provided by Advanced Traveler Information Systems (ATIS) and their previous experience. The availability and perceived quality of traffic information are explicitly modeled within the proposed framework. The uncertainty associated with a drivers travel time estimate is modeled in a stochastic dynamic framework and is incorporated in a travel choice model. Each driver uses a disutility function of perceived travel time and perceived schedule delay to evaluate the alternative travel choices, then selects an alternative based on the utility maximization principle. The perception updating model and the choice model are integrated with a dynamic traffic simulator (DYNASMART). Empirical results from the simulation experiments and their implications are also presented.

Robust improvement schemes for road networks under demand uncertainty

This paper is concerned with development of improvement schemes for road networks under future travel demand uncertainty. Three optimization models, sensitivity-based, scenario-based and min-max, are proposed for determining robust optimal improvement schemes that make system performance insensitive to realizations of uncertain demands or allow the system to perform better against the worst-case demand scenario. Numerical examples and simulation tests are presented to demonstrate and validate the proposed models.

OPTIMAL INSPECTION AND MAINTENANCE POLICIES FOR INFRASTRUCTURE NETWORKS

State-of-the-art infrastructure management systems use Markov decision processes (MDPs) as a methodology for maintenance and rehabilitation (M&R) decisionmaking. The underlying assumption in this methodology is that inspections are performed at pre-set and fixed time intervals and that they reveal the true condition of the facility, with no measurement error. As a result, after an inspection, the decisionmaker can apply the activity prescribed by the optimal policy for that condition state of the facility. In prior research, the second author of this paper has applied a methodology for M&R activity selection accounting for the presence of both forecasting and measurement uncertainty--the latent Markov decision process (LMDP), an extension of the traditional MDP that relaxes the assumptions of error-free annual facility inspections. In this paper, the authors extend this methodology to include network-level constraints. This can be achieved by extending the LMDP model to the network-level problem through the use of randomized policies. Both finite- and infinite-horizon formulations of the network-level LMDP are presented. A case study application demonstrates the expected savings in life cycle costs that result from increasing the measurement accuracy used in facility inspections and from optimal scheduling of inspections.

A STEADY-STATE SOLUTION FOR THE OPTIMAL PAVEMENT RESURFACING PROBLEM

This paper presents a solution approach for the problem of optimising the frequency and intensity of pavement resurfacing, under steady-state conditions. If the pavement deterioration and improvement models are deterministic and follow the Markov property, it is possible to develop a simple but exact solution method. This method removes the need to solve the problem as an optimal control problem, which had been the focus of previous research in this area. The key to our approach is the realisation that, at optimality, the system enters the steady state at the time of the first resurfacing. The optimal resurfacing strategy is to define a minimum serviceability level (or maximum roughness level), and whenever the pavement deteriorates to that level, to resurface with a fixed intensity. The optimal strategy does not depend on the initial condition of the pavement, as long as the initial condition is better than the condition that triggers resurfacing. This observation allows us to use a simple solution method. We apply this solution procedure to a case study, using data obtained from the literature. The results indicate that the discounted lifetime cost is not very sensitive to cycle time. What matters most is the best achievable roughness level. The minimum serviceability level strategy is robust in that when there is uncertainty in the deterioration process, the optimal condition that triggers resurfacing is not significantly changed.

Optimal maintenance and repair policies in infrastructure management under uncertain facility deterioration rates: an adaptive control approach

In this paper we present two adaptive control formulations that explicitly include uncertainty in characterizing a facilitys deterioration rate in the process of developing maintenance and repair policies. The formulations use condition information that stems from the operation of a facility to improve the characterization over a finite planning horizon. We discuss issues related to the implementation and computational complexity of the formulations. Through a computational study, we show that the economic benefits can be achieved by implementing the adaptive control formulations. These benefits are most significant in situations where the initial expected deterioration rate of the facility is not a good representation of its actual deterioration rate.

USING DURATION MODELS TO ANALYZE EXPERIMENTAL PAVEMENT FAILURE DATA

Predicting pavement performance under the combined action of traffic and the environment provides valuable information to a highway agency. The estimation of the time at which the pavement conditions will fall below an acceptable level (failure) is essential to program maintenance and rehabilitation works and for budgetary purposes. However, the failure time of a pavement is a variable event; terminal conditions will be reached at different times at various locations along a homogeneous pavement section. A common problem in modeling event duration is caused by unobserved failure events in a typical data set. Data collection surveys are usually of limited length. Thus, some pavement sections will have already failed by the day the survey starts; others will reach terminal conditions during the survey period, whereas others will only fail after the survey has been concluded. If only the failure events observed during the survey are included in the statistical analysis (disregarding the information on the events after and before the survey), the model developed will suffer from truncation bias. If the censoring of the failure events is not accounted for properly, the model may suffer from censoring bias. An analysis of the data collected during the Road Test sponsored by the American Association of State Highway Officials (AASHO) is presented. The analysis is based on the use of probabilistic duration modeling techniques. Duration models enable the stochastic nature of pavement failure time to be evaluated as well as censored data to be incorporated in the statistical estimation of the model parameters. The results, based on sound statistical principles, show that the failure times predicted with the model match the observed pavement failure data better than those from the original AASHO equation.

Pavement Resurfacing Policy for Minimization of Life-Cycle Costs and Greenhouse Gas Emissions

In recent decades, pavement management optimization has been designed with the objective of minimizing user and agency costs. However, recent analyses indicate that pavement management decisions also have significant impacts on life-cycle greenhouse gas (GHG) emissions. This study expands beyond minimization of life-cycle costs to also include GHG emissions. Previous work on the single-facility, continuous-state, continuous-time optimal pavement resurfacing problem is extended to solve the multicriteria optimization problem with the two objectives of minimizing costs and GHG emissions. Results indicate that there is a trade-off between costs and emissions when developing a pavement resurfacing policy, providing a range of GHG emissions reduction cost-effectiveness options. Case studies for an arterial and a major highway are presented to highlight the contrast between policy decisions for various pavement and vehicle technologies.

EFFECT OF PERFORMANCE MODEL ACCURACY ON OPTIMAL PAVEMENT DESIGN

In the first part of this paper, an analysis of the data collected during the American Association of State Highway Officials (AASHO) Road Test, based on probabilistic duration modeling techniques, is presented. Duration techniques enable the stochastic nature of pavement failure time to be evaluated as well as censored data to be incorporated in the statistical estimation of the model parameters. The second part of this paper presents the use of economic optimization principles for determining the optimal design of flexible pavements. We study the effect of deterioration model accuracy on optimal design and lifecycle costs, by comparing three models. The first is a simple regression model developed by the AASHO, which forms the basis of design standards in use today. The second is a regression model that was developed with the same AASHO data set, but that includes a correction for data censoring. The third model is the probabilistic model developed in the first part of this paper. The results show that the AASHO model, when used as an input to lifecycle cost minimization, produces a pavement structural number that is lower than that produced by using the other two deterioration models. This results in shorter pavement lives and higher costs due to more frequent resurfacing. The savings in lifecycle cost accrued by using optimal structural number are shown to be quite significant, offering a sound basis for revising current design practices.

Reliability-Based System-Level Optimization of Bridge Maintenance and Replacement Decisions

This paper addresses the problem of optimizing bridge maintenance and replacement (M&R) decisions for a heterogeneous system of facilities. The objective is to determine optimal M&R policies for each facility over a finite planning horizon based on the knowledge of the current facilities conditions and on the prediction of future conditions. The system-level problem is based on the results of an M&R optimization problem for each facility. The results of the facility-level optimization are incorporated in a reliability-based, bottom-up, system-level formulation that provides recommendations for each individual facility. We derive sufficient conditions for optimality and prove the result for the continuous case. A parametric study shows that the results obtained in the discrete-case implementation of the solution are valid approximations of the continuous case results. The computational efficiency of the system-level solution makes the formulation suitable for systems of realistic sizes.