Impact of the Structural Defects on Risk Assessment of Concrete Bridges According to the Italian Guidelines 2020

Abstract

The Italian infrastructure network of roads and bridges is one of the most complex in the world due to the territory orography. Italy is strongly interested in seismic and hydrogeological hazards, and, in addition, degradation and obsolescence phenomena are common in infrastructures nowadays approaching the end of their nominal life. Furthermore, these infrastructures are subjected to continuous traffic load increase over time. In 2020, the Italian Ministry of Infrastructure and Transport (MIT) published the guidelines for risk classification and management, safety assessment, and monitoring of existing bridges (LG2020) as an attempt to unify the multiple procedures of inspection, monitoring, and maintenance of infrastructures. The multilevel approach proposed in the Italian guidelines for the management of the complex existing system of bridges is herein discussed and investigated, focusing on an operational methodology to evaluate the impact of structural defects on the risk assessment. This study aims to develop an operational methodology for the application of the procedure generically depicted in the LG2020 for the attribution of the level of defectiveness based on the outcomes of the periodical inspections. In particular, such a methodology is applied to two of the most widespread bridge structural typologies in the Mediterranean area: reinforced concrete (RC) and prestressed RC (PRC) bridges. The defects’ extent and level to structural members are associated with the proposed procedure for different bridge risk ratings. The work presents a useful tool to proceed from the outcomes of the inspections to the assignment of a level of defectiveness for the bridge, which enters into the risk assessment. This is to drive decision-makers in the definition of future actions and interventions, such as the detailed assessment of safety level and relevant strengthening interventions or installation of continuous monitoring systems.

1. Introduction

Italy has an articulated road system developing for 840,000 km, grouped into highways, state, regional, provincial, and municipal roads; the presence of one bridge for each 2 km of road is estimated [1]. An infrastructure network represents a focal point of collective interest. Indeed, most commercial, industrial, and relief activities are carried out through the road system [2] in the presence of natural disasters.

A total of 45% of the management of Italian roads belongs to a state company, 20% belongs to road and motorway concessionaires, and the remaining 35% belongs to local authorities [2]. This diversification from an administrative point of view also makes difficult the standardization of the entire management and maintenance process [3].

In the Mediterranean region, most of the infrastructures were built after World War II, during the so-called “economic boom”; reinforced concrete (RC) and prestressed reinforced concrete (PRC) structures are the most diffused ones. The use of PRC reduced the construction time, guaranteeing a higher resistance and the possibility of covering considerable spans; nevertheless, constructions in PRC are currently characterized by considerable problems, especially in post-tensioned systems [4]. PRC bridges were built mostly following elementary schemes, with no particular emphasis on ductility [5]. Moreover, they were built when, at least in Italy, there was not yet wide knowledge of prestressing system technologies. Several recent bridge collapses pointed to the urgent need to install monitoring systems to control safety conditions and deterioration evolution, especially of prestressed concrete bridges. Existing bridges that have made Italian engineering internationally famous are currently affected by signals of degradation or damage, generally referred to in the following as “defects”. They may be related to different reasons, such as an increase in traffic loads, concrete degradation, lack of maintenance, sudden events (such as earthquakes or hurricanes [6,7]), and climate change [8,9]. Also, the environmental exposure conditions strongly affect infrastructure durability. For example, chloride attacks in coastal areas and those resulting from salt spreading on bridges to prevent ice formation represent a trigger for RC cracking and reinforcement corrosion [10,11,12]. Moreover, bridge collapses that occurred in recent years highlighted the inevitable need to focus on the management and maintenance of existing infrastructure networks.

The high vulnerability and strong degradation level of existing bridges worldwide have been discussed by several recent studies [6,8,9,13]; the presence of defects on the main elements of bridges, such as bearings, piers, and abutments, have been identified as critical for their performance [9]. The degradation control is crucial for optimizing maintenance strategies from an economic and management perspective. Preventive maintenance on bridge members is a key aspect of maintaining the operation of infrastructure [6], such as the programming interventions on existing bridges [14,15]. For this reason, inspections and continuous monitoring are crucial tools to prioritize interventions and define timelines [6]. The presence of defects and their levels are generally denoted by scales expressed with qualitative adjectives, such as “high”, “medium”, and “low”.

In line with the international context, in 2020, the Italian Ministry of Infrastructure and Transport (MIT) published the guidelines for risk classification and management, safety assessment, and monitoring of existing bridges (LG2020, [16]) and their operative instructions [17], as an attempt to unify the multiple procedures of inspection, monitoring, and maintenance of infrastructures. These documents provide indications and methodologies for the management of existing bridges, applicable by administrative entities at national and local (i.e., provincial or municipal) levels [3,18]. The LG2020 adopts a multilevel approach in which the risk assessment is based on structural, seismic, and hydrogeological aspects. The approach leads to the definition of the bridge Warning Class; to this aim, the number and weight of the defects are crucial parameters. The Warning Class drives the decisions on the future interventions to be carried out (e.g., monitoring and safety assessment). Limited-in-time inspection procedures can be complemented with continuous monitoring systems, which can be consulted remotely in real-time and automated [19,20,21,22,23]. This study aims to develop an operational methodology for the application of the procedure generically depicted in the LG2020 for the attribution of the level of defectiveness based on the outcomes of the periodical inspections. The motivation for this study comes from the need to rattle off some processes that in the LG2020 are mentioned but not well explained in detail.

The methodology is applied to two very common existing bridge typologies, RC and PRC bridges. Some case studies affected by different defect types, extent, and levels were selected to illustrate the methodology application. The defects classified by means of LG2020 provisions on these real case studies are used to experiment with the multilevel approach proposed in the LG2020 for the management of the complex existing system of bridges. Particular attention is devoted to defects on “critical” elements; these defects present features of potential fragile failure that can involve the entire bridge. Clearly, these types of defects need to be contextualized inside the structural mechanisms that characterize the bridge. This process is detailed in this work. Different types of degradation phenomena are analyzed on RC elements and PRC elements such as prestressing cables, bearings, beam ends, and piers. The goal of this work is to provide a useful tool to proceed from the outcomes of the inspections to the assignment of a level of defectiveness for the bridge, which enters into the risk assessment. The potential and the application procedure of the methodology are illustrated through the analysis of applicative cases according to LG2020.

2. Methodology

This section presents the quantification of the level of defectiveness based on the outcomes of the inspections. Firstly, a summary of the most important defects in RC and PRC structural elements is presented (Section 2.1). Then, the level of defectiveness is discussed within the context of the Warning Classes proposed by the LG2020 (Section 2.2). Finally, the procedure for the assignment of the level of defectiveness is proposed (Section 2.3). The proposed methodology can be used worldwide, since the analysis of the defects is independent of the regulations in force in each country.

2.1. Defects on RC and PRC Structural Elements

The section examines typical defects that may occur on structural elements of RC and PRC bridges.

Table 1. Defects on RC and PRC bridge elements.

Defect	Elements
Corroded reinforcement	RC decks, PRC decks, Gerber saddles, Piers, Abutments
Crush cracks	Supports areas, Piers, Abutments
Diagonal or transversal cracks	RC decks, PRC decks, Gerber saddles, Piers, Abutments
Prestressed reinforcement section reduction	PRC decks
External prestressed reinforcement load reduction	PRC decks
Deformed longitudinal reinforcement	RC decks, PRC decks, Gerber saddles, Piers, Abutments
Out of plumb	Piers, Abutments
Stirrups break	RC decks, PRC decks, Gerber saddles, Piers
Washed away/damaged beam ends	RC decks
Degraded ducts and/or oxidized wires	PRC decks
Anchorage bar leakage	PRC decks

synthetizes the most important defects commonly detected on RC and PRC bridges, specifying on which element that defect can occur.

Several studies on the deterioration of RC and PRC structural elements of bridges have been conducted both in Italy and abroad. The most widespread are cracking of concrete and corrosion of reinforcement [8].

2.2. Classification of Defectiveness within the Warning Classes Proposed by LG2020

The multilevel approach proposed by the LG2020 is developed on six levels of analysis (from Level 0 to Level 5). Level 0 provides the census of all the main bridge characteristics through the collection of available information and documents. Then, Level 1 provides the execution of visual inspections and rapid surveys aiming to evaluate the bridge’s geometry, structural features, and degradation state (reported on the inspection defect sheet), as well as the geomorphological and hydraulic characteristics of the area. The Warning Class of the bridge is determined at Level 2 based on hazard, vulnerability, and exposure parameters. Levels 3 and 4 involve the static and seismic assessment of the bridge with different levels of accuracy. Finally, Level 5 involves the resilience analysis of the road network to which the considered bridges belong. The crucial point of the multilevel approach is the determination of the Warning Class, which is the result of the combination of four attention sub-classes related, respectively, to structural and foundational, seismic, landslides, and hydraulic issues.

This work focuses on the study of the level of defectiveness included in the Structural and Foundational Warning Class for typical situations of bridge defectiveness. The adopted methodology is explained in detail in Section 2.3.

2.3. Procedure for the Assignment of the Level of Defectiveness

The defectiveness is a crucial parameter for the calculation of the Warning Class. For example, a high level of defectiveness means a high Structural and Foundational Warning Class, which means a high overall Warning Class.

The proposed procedure to derive the level of defectiveness for a generic bridge is shown in Figure 1.

Firstly, to have a correct knowledge of the importance of the single defects in the potential reduction in the bridge capacity, it is important to know the main structural features of the bridge (e.g., geometry and structural typology). Then, the inspection reports related to the bridge defects can be analyzed. In fact, according to the LG2020 and the related Operative Instructions, the level of defectiveness is linked to the current state of conservation of the structure, which can be assessed by analyzing the outcomes of the defect sheets produced inside the Level 1 activities. An excerpt of a defect sheet (e.g., for the concrete piers) provided in the LG2020 to be filled in by an inspector is reported in Figure 2. In the sheet, G is the parameter indicating the severity of the detected defect (varying between 1 and 5, from low to high), “PS” indicates the possibility that the defect affects the static behavior of the bridge, and “NA”, “NR”, and “NP” stand, respectively, for “not applicable”, “not detectable”, and “not present”.

The LG2020 provides 5 levels of defectiveness (from low to high) according to severity (G), extent (k₁), and intensity (k₂) of the detected defects. Moreover, the type of structural element and its incidence on the overall structural behavior of the bridge drives the assignment of the level of defectiveness.

Figure 1 shows that considerations about the presence of defects on “critical elements” or defects leading to a “critical condition” are crucial in the procedure. The Operative Instructions of the LG2020 define an element whose crisis can lead to the crisis of the whole structure or part of it as a “critical element” (e.g., Gerber saddles, bearings, prestressing cables), and a condition of a possible collapse caused by the presence of defects with high or medium-high severity (G = 5 or G = 4, respectively) and high intensity and extent on a considerable number of elements in terms of number and/or position as a “critical condition”. Possible critical conditions are, for example, ongoing kinematics and crack patterns with high extent and intensity. If at least one defect of severity G = 4 or G = 5 and whatever intensity is present in a critical element, consequently, the level of defectiveness of the bridge is assumed to be high. It is important to focus not only on the severity, intensity, and extent of defects usually reported in the LG2020 inspection defects but also on the distribution of the defects on the elements. Locating each defect on the plan distribution gives stakeholders and managing institutions a complete picture of the situation. The graphical representation of the defects allows us to have a picture of the generally prevalent defect typologies on RC and PRC bridge, their intensity, and spatial distribution, thus arriving at an overall judgment on the static properties that influence the definition of the level of defectiveness. Sometimes, severe defects are scattered over the entire bridge plan, and sometimes, a high concentration can be found in a limited area. Therefore, the derivation of the defect level cannot be separated from a correct identification in plan and/or elevation. Note that, for each span, if at least 50% of the elements are characterized by a medium defectiveness level, the overall defectiveness of the span can be assumed to be medium; otherwise, it can be assumed to be between medium-low and low, depending on the most common one.

The flowchart presented in Figure 1 explains all the conceptual steps to derive the level of defectiveness from the list of defects present in a viaduct. Starting from the LG2020, eight potential defectivity conditions can be distinguished, described in

Table 2. Description of the defects categories.

Case	Description
1	G = 4 or G = 5 defects on a critical element with whatever intensity
2	More than one G = 4 or G = 5 defect with high intensity on element whose failure can affect the static behavior of the bridge: critical condition
3	More than one G = 4 or G = 5 defect with high intensity on element whose failure can affect the static behavior of the bridge: no critical condition
4	One G = 4 or G = 5 defect with high intensity on element whose failure can affect the static behavior of the bridge
5	G = 4 or G = 5 defects with high intensity on element whose failure cannot affect the static behavior of the bridge, and G = 5 defects with medium-low intensity
6	G = 4 defects with medium-low intensity
7	G = 1, G = 2, G = 3 defects (number of defects/number of elements ≥ 50%)
8	G = 1, G = 2, G = 3 defects (number of defects/number of elements < 50%)

.

It could be of interest to assign a weight to each condition, for example, by following an exponential law, with decreasing severity from case 1 to case 8. In this way, an effort with the aim to quantify the level of defectiveness could be made. However, this is not easy to perform, since the estimation of the defectiveness condition of each bridge should be analyzed case by case. Moreover, a detailed definition of the “critical condition” should be given by the LG2020, for example, as a function of the number of defects that affect the static properties and the number of degraded elements. Another important aspect is the way to pass from the defectiveness of the single span and of each of the sub-structure groups (e.g., pier and its supports) to the overall bridge defectiveness level. For the entire bridge, the maximum defectiveness level among the ones found on the spans and each sub-structure group should be assigned. All these aspects were evaluated in the applications.

3. Applications

The proposed procedure is applied to five spans belonging to existing RC bridges. The spans were chosen to take into account most of the defects listed in Table 1, e.g., corroded reinforcement, crush cracks, diagonal or transversal cracks, prestressed reinforcement section reduction, and external prestressed reinforcement load reduction. Note that the applications shown in the following are just methodological and can be repeated for any bridge, independently from the specific defects.

3.1. Span 1 and Span 2

The first two case study spans (Span 1 and Span 2) belong to a multiple-span PRC (post-tensioned) bridge. The analyzed spans—about 35.00 m in length—present an isostatic scheme, with Gerber saddles supporting other spans. The decks consist of five girders connected by eleven transversal transoms and two end Gerber saddles, as well as thick upper and lower slabs. The bearings are made of neoprene. The bridge piers are multi-frame, with single columns connected by transversal beams. The longitudinal view and the transversal section in the middle of the case study bridge are reported in Figure 3.

The inspection sheet of the bridge is analyzed. Regarding Span 1, among other defects detected on the deck, corrosion of reinforcement was detected on the Gerber saddles. Then, a defect with severity G = 5 is on a critical element. For this reason, the Gerber saddles fall in “Case 1”, with a “high” defectiveness level. By collecting the defectivity level of all the elements of the deck, the following statistics are determined: 40% of elements have “low” defectivity, 30% have a “medium-low” defectivity, 10% have a “medium” defectivity, 10% have a “medium-high” defectivity, and the remaining 10% have a “high” defectivity. Thus, according to the LG2020, the level of structural and foundational defectivity of this case study span is “high”. Regarding Span 2, a critical condition is present. In fact, three girders out of five present corroded reinforcement with high intensity. The defectivity levels of all the elements are as follows: 40% of elements have a “low” defectivity, 20% have a “medium-low” defectivity, 10% have a “medium” defectivity, 20% have a “medium-high” defectivity and the remaining 10% have a “high” defectivity. According to the LG2020, the level of structural and foundational defectivity of this case study deck is “high”.

To deepen the motivations of this result, the main issue is related to the corrosion of reinforcement, which consists of the advanced degradation of steel bars embedded in the concrete following the initial oxidation phase, which often leads to a reduction in its cross-sectional area. In the most advanced stage, corrosion of reinforcement also induces a detachment of the concrete cover, so the corroded reinforcement is visible. This defect can occur because of the chemical and physical phenomena induced by aggressive agents to which the reinforcement cover is exposed. The external concrete cover is generally subjected to carbonation. Carbonation occurs when calcium hydroxide and carbon dioxide in the air form calcium carbonate. The latter, due to its high acid content, lowers the pH of the concrete and removes the oxide film around the reinforcement, which begins to oxidize. Another cause of reinforcement deterioration is bad water regimentation on the deck, which induces a percolation of rainwater, which infiltrates into the areas of discontinuity between one deck and another. Finally, a chloride attack can affect RC elements. It is one of the causes of degradation most extensively investigated in the literature (e.g., [6,9]). It is due to the proximity of bridges to coastal areas or salt spreading to prevent ice formation. When the chloride concentration exceeds a certain critical threshold, the steel depassivates, and corrosion of the reinforcement begins and then has a subsequent propagation. In addition to these chemical–physical phenomena, climatic changes, such as annual temperature and relative humidity, can increase the deterioration process [9]. Reinforcement corrosion is a typical defect of RC and PRC bridge decks, Gerber saddles, piers, and abutments. As an example, the configuration of the corroded reinforcement on the Gerber saddle of the case study Span 1 is shown in Figure 4. Such a defect leads to a reduction in the strut and tie mechanism capacity of the saddle, especially on the concrete side, for the reduction in the section of the strut.

3.2. Span 3 and Span 4

Span 3 and Span 4 are part of a bridge consisting of two carriageways, each of which is composed of five spans, for a total length of about 110.00 m. The decks of four spans are composed of simply supported PRC beams with wires connected by transoms and a thick slab. The decks of the last span are formed by simply supported precast PRC beams and transoms with post-tensioned cables with a thick slab. The width of the decks is about 13.00 m in rectum. Two piers over four consist of three circular section stems connected by a pulvinus at the top. Instead, the other two piers consist of two rectangular section stems connected at the top by a pulvinus. The bearings are made of steel and vulcanized neoprene. The longitudinal views and the plans of the case study Span 3 and Span 4 are reported in Figure 5a,b, respectively.

The inspection sheet of the bridge shows that for Span 3, 63% of elements have a “low” defectivity, 25% have a “medium-low” defectivity, and the remaining 13% have a “medium” defectivity. Thus, according to the LG2020, the level of structural and foundational defectivity of the case study Span 3 is “low”.

Span 4 has a significant number of defects: 50% of elements have a “low” defectivity, 38% have a “medium” defectivity, and the remaining 13% have a “medium-high” defectivity. This leads, according to the LG2020, to the level of structural and foundational defectivity of Span 4 “medium-high”.

The main issue is related to the crush cracks observed on several RC supports, one of them with high intensity. The crush cracks are generally inclined at 45°. They are probably due to an excess of compressive stresses since, in this part of the structure, loads are transferred from the decks to the structures in elevation. The formation of crush cracks can derive either from uncorrected design in terms of geometry and/or reinforcements or poor quality of the concrete. Moreover, in simply supported decks, the discontinuities present between decks often represent a preferential path for the infiltration of water and aggressive agents that can cause damage. Such cracks are generally coupled; they generate a wedge of material that tends to pull apart. A focus on an RC support affected by crush cracks is reported in Figure 6.

3.3. Span 5

Span 5 belongs to a bridge consisting of two carriageways, each of which is composed of seven spans, for a total length of about 313.00 m. The decks are composed of simply supported PRC beams with post-tensioned cables connected by transoms and a thick slab. The width of the decks is about 10.00 m. The piers have a single column. The bearings are made of steel. The longitudinal view and the plan of the case study Span 5 are reported in Figure 7.

The inspection sheet of the bridge shows that, among other defects detected on the deck, prestressed reinforcement section reduction was detected. Then, a defect with severity G = 5 is on the prestressed cable, which is a critical element. For this reason, it falls into “Case 1”, with a “high” defectiveness level. The defectivity levels of all the elements of the deck are as follows: 20% of elements have a “low” defectivity, 50% have a “medium-low” defectivity, 10% have a “medium” defectivity, 10% have a “medium-high” defectivity, and the remaining 10% have a “high” defectivity. Thus, according to the LG2020, the level of structural and foundational defectivity of this case study deck is “high”.

The prestressed reinforcement section reduction may be due to chemical and physical causes, such as chloride attack, carbonation, and poor water regimentation. In particular, water infiltrations in the sheaths of the prestressing cables represent the main trigger of the cables’ corrosion. An advanced stage of corrosion can lead to a reinforcement section reduction and, thus, the ability of the involved element to withstand a certain stress level. Details related to such defects are reported in Figure 8.

4. Discussion and Proposal for the Resilience Improvement

The level of defectiveness is crucial in the definition of the Warning Class. Indeed, a “high” level of defectiveness induces a “high” Vulnerability Class, regardless of any other parameter. Then, a “high” vulnerability class induces a “high” Structural and Foundational Warning Class, independently of the Hazard Class and the Exposure Class. Finally, this ends with the attribution of a “high” total Warning Class, whatever the seismic Warning Class and the hydraulic and landslide Warning Classes are. However, in this study, the influence of the defects on the Structural and Foundational Warning Class, which is just a component of the full procedure to derive the Warning Class of a bridge, was deeply investigated.

While it is acknowledged that the level of structural defects holds a prominent role in assigning the “high” Warning Class, it is worth mentioning that it is not the only element that can lead to this result. For example, another element that can influence a “high” Warning Class is the construction period, which dictates degradation speed and design code class.

For bridges with a “high” total Warning Class, accurate assessments should immediately be carried out according to the LG2020. It is necessary to provide the execution of ordinary or extraordinary periodical inspections, as well as the installation of monitoring systems. Instead, a “medium-high” level of defectiveness is not binding on the definition of the total Warning Class. It could induce a vulnerability class ranging between “medium-low” and “high”. This can lead to any level of Structural and Foundational Warning Class and, finally, to any level of total Warning Class, which will depend on all the other parameters described in the LG2020. In RC and PRC bridges, the most widespread defect is the presence of cracks in concrete. Ordinary and prestressed RC decks, Gerber saddles, piers, and abutments can be generally affected by light cracks, which appear when the loads induce tensions exceeding the concrete tensile strength. Transversal cracks usually develop along the transverse axis of elements such as beams and transoms. They can have multiple causes (e.g., design errors, lack of transverse reinforcement). Instead, diagonal cracks on vertical elements, such as piers and abutments, are due to foundation problems. In horizontal elements, the presence of diagonal cracks may be related to an overload or a wrong structural behavior and represents a severe defect to be considered. When cracks are identified in the support areas of the beams, they can be attributable to shear stress, which is dangerous since it leads to a fragile crisis mechanism [24]. It is a note that could be explicitly considered in a future update of the LG2020, such as the time-dependent prestress losses in concrete bridges, which are mainly due to the phenomena of tendon relaxation, creep, curing, cracking, and shrinkage of concrete [25,26].

One of the greatest challenges for management companies is to ensure that an infrastructure system is not only efficient but also resilient. The resilience of a system is the capacity to absorb and recover functionality loss from damages and to adapt to new situations. This goal can be pursued by providing careful planning and then taking into account the system’s response to the evolution of degradation and/or any damage following the occurrence of some event. After that, interventions should be implemented to restore functionality and, in addition, take into account future stress factors, such as permanent exposure to climate change and the increase in variable traffic loads [27]. When the Warning Class is medium-high/high, one of the most effective ways to improve the resilience of infrastructures is to provide not only periodic visual inspections and special inspections but, above all, an active Structural Health Monitoring (SHM) system [19,20]. This is also suggested by the LG2020. It consists of continuous monitoring, useful both for making timely decisions (e.g., limiting or interrupting the traffic) and for long-term planning surveillance activities. It is crucial to correctly manage risk control and mitigation activities, reducing management and maintenance times and costs and aiming, therefore, at a sustainable development of the entire system. Sensors can be employed for the continuous monitoring systems of RC bridges. The most used are inclinometers (monitoring of rotations), strain gauges (monitoring of deformations), accelerometers (dynamic monitoring), and LVDTs (Linear Variable Displacement Transducers, monitoring of linear displacements). In addition to the abovementioned sensors, in recent years, new satellite monitoring techniques have been proliferating, especially to supervise large areas. In the last 20 years, for example, DInSAR (Differential Interferometric Synthetic Aperture Radar) techniques have been widely developed to monitor surface deformations in urban areas, both at local and territorial levels [21,22,23,28]. In addition, regardless of continuous digital monitoring, to further investigate the state of deterioration, the LG2020 requires so-called special inspections, especially for PRC bridges with post-tensioned cables, considered particularly critical structures. These types of inspections include a plan of non-destructive or semi-destructive investigations, such as pacometric, ultrasonic, endoscopic, and tomographic. They are essential for identifying the path of the cables and their stress state and defining the location of possible voids or defects.

5. Conclusions

This study investigates and discusses the aspects related to the influence of the level of defectiveness, included in the multilevel approach proposed in the Italian guidelines for risk classification and management, safety assessment, and monitoring of existing bridges (LG2020), for the classification and risk management of existing bridges. Starting from the analysis of common defects typically found on RC and PRC bridges, a procedure to assign the bridge level of defectiveness has been presented. It depends on a deep analysis of the defects reported in the inspection sheets, taking into account their diffusion and spatial distribution. The study reports a focus on the defects with high and medium-high severity, crucial for the attribution of the level of defectiveness of the bridge. In such a case, the presence of defects in critical elements plays a key role. According to the LG2020, a defect with high or medium-high severity on a critical element, regardless of its intensity, makes the level of defectiveness “high”. Five case study spans, affected by different kinds of defects and various distributions, have been presented to investigate the defects in critical elements typical of RC and PRC bridges. The analysis was extended to all the defects noticed in the inspection sheets to show how the presence of defects with high or medium-high severity and high intensity and extent on a significant number of elements may lead to a “high” level of defectiveness.

The study has shown, by means of real case studies, a methodology for the attribution of a bridge’s level of defectiveness based on the outcomes of the periodical inspections. Differently from almost all the other parameters that influence the preliminary risk assessment (e.g., the construction period and the geometry), defectiveness is a parameter that can evolve in time: for better, if maintenance interventions are carried out, or for worse, if no actions are performed in time. Then, a clear process to quantify the defects, step by step, can be very useful for the definition of the partial Warning Classes and for the final Warning Class of a bridge. Moreover, the managing Company of a bridge network can use this procedure to perform different Warning Class scenarios by varying the defectiveness level in order to understand which bridge of the network should be prioritized for reparation. Then, it represents a useful tool to drive decision makers from the outcomes of in situ inspections to the definition of future actions and interventions to optimize the management and maintenance of existing infrastructure networks.