Johan Sverdrup: the Final Lap of a True North Sea Giant

The second phase of Johan Sverdrup came on stream in December 2022. This paper focuses on the execution of Johan Sverdrup phase 2 and describes the assessments and investments for improved oil recovery (IOR) from one of the largest oil fields in Norway. The Johan Sverdrup field development has been called Equinor's ‘digital flagship', and this paper includes the proof of concept for the digital initiatives after more than three years of operation. Despite the Covid-19 pandemic Johan Sverdrup phase 2 has been able to deliver on schedule, under budget, and with an excellent safety record. The paper includes experiences from the concept development and engineering phase to the global contracting strategy, through the construction on multiple building sites in Norway and globally, and until the end of the completion phase offshore Norway. Johan Sverdrup is the third largest oil field on the Norwegian Continental Shelf (NCS), and with recoverable reserves estimated at 2.7 billion barrels of oil equivalents, has the resources to be a North Sea Giant. Start-up of the Johan Sverdrup phase 2 extends and accelerates oil and gas production from the NCS for another five decades. This paper aims to highlight what it took to make Johan Sverdrup a true North Sea Giant, fit for the 21st century: a safe and successful execution of a mega-project, with next-generation facilities adapted to a more digital way of working, with an ambition to profitably recover more than 70% of the resources, while limiting carbon emissions from production to a minimum. In many ways the Johan Sverdrup development has set a new standard for project execution in Equinor. The impact of different variables made during the execution of the project, such as the Covid-19 pandemic, market effects, procurement strategies, value improvement initiatives, execution performance and reservoir characteristics is addressed, as well as describing assessments and investments for improved oil recovery (IOR). Data acquisition, Permanent Reservoir Monitoring (PRM), fibre-optic monitoring of wells, innovative technologies, and digitalization, as well as new ways of working are included. Equinor ´s digital strategy was established in 2017, and Johan Sverdrup was highlighted as a digital flagship at that time and a frontrunner in applying digital solutions to improve safety and efficiency from the development to the operational phase. What has been implemented so far together with experiences will be shared. © 2023, Offshore Technology Conference.

Embracing Remote Technology in the System Safety World

The advent of the Covid-19 pandemic that began at the end of 2019 accelerated technology to support remote work environments. While the technologies were not new, the capabilities of those technologies significantly increased. The number of users embracing the remote working technologies significantly increased within a very short timeframe. The expanded remote work capabilities have enabled new collaborating mechanisms that will carry-forth in the future, pandemic or not. SAE ARP 4761 [1] provides guidelines to perform a safety assessment. The Zonal Safety Assessment (ZSA) is one of the tools described in ARP 4761. An aspect of ZSA is an audit (or inspection) of the physical article. The remote work capabilities, along with cameras and software presenting a virtual environment, can allow individuals to participate in the physical audit without traveling to the site of the physical article. ©2023 SAE International. All Rights Reserved.

PFL: A Probabilistic Logic for Fault Trees

Safety-critical infrastructures must operate in a safe and reliable way. Fault tree analysis is a widespread method used for risk assessment of these systems: fault trees (FTs) are required by, e.g., the Federal Aviation Administration and the Nuclear Regulatory Commission. In spite of their popularity, little work has been done on formulating structural queries about and analyzing these, e.g., when evaluating potential scenarios, and to give practitioners instruments to formulate queries on in an understandable yet powerful way. In this paper, we aim to fill this gap by extending [37], a logic that reasons about Boolean. To do so, we introduce a Probabilistic Fault tree Logic is a simple, yet expressive logic that supports easier formulation of complex scenarios and specification of FT properties that comprise probabilities. Alongside, we present, a domain specific language to further ease property specification. We showcase and by applying them to a COVID-19 related FT and to a FT for an oil/gas pipeline. Finally, we present theory and model checking algorithms based on binary decision diagrams (BDDs). © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

The Next Chapter in Global Compliance Auditing

The recent COVID-19 pandemic spurred evolutionary changes to how risk management programs are executed across the globe. As companies responded to the effects the virus had on their operations, they also had to continue their compliance verifications despite the inability of audit teams to visit sites for a first-hand evaluation of program status. Audit teams were driven to innovate to accomplish their mission. This metamorphosis has forever changed the process by which compliance auditing is conducted. These methodologies continue to evolve as the pandemic endures. This article will discuss the author's views of advancements in compliance auditing as a direct result of the pandemic. It will feature tools that have been developed, including planning tips, audit strategies, auditor utilization, and various concepts that audit teams can employ. The author will explain how she believes these aspects will affect the future of post-pandemic compliance auditing, particularly focusing on auditing for Process Safety Management (PSM) system requirements. © 2022 AIChE Spring Meeting and 18th Global Congress on Process Safety, GCPS 2022. All rights reserved.

PFL: A Probabilistic Logic for Fault Trees

Safety-critical infrastructures must operate in a safe and reliable way. Fault tree analysis is a widespread method used for risk assessment of these systems: fault trees (FTs) are required by, e.g., the Federal Aviation Administration and the Nuclear Regulatory Commission. In spite of their popularity, little work has been done on formulating structural queries about and analyzing these, e.g., when evaluating potential scenarios, and to give practitioners instruments to formulate queries on in an understandable yet powerful way. In this paper, we aim to fill this gap by extending [37], a logic that reasons about Boolean. To do so, we introduce a Probabilistic Fault tree Logic is a simple, yet expressive logic that supports easier formulation of complex scenarios and specification of FT properties that comprise probabilities. Alongside, we present, a domain specific language to further ease property specification. We showcase and by applying them to a COVID-19 related FT and to a FT for an oil/gas pipeline. Finally, we present theory and model checking algorithms based on binary decision diagrams (BDDs). © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Autonomous Medical Assistive Robot for Medical Camp

The outbreak of the Covid-19 epidemic has devastated the generation and impacted multiple layers of the healthcare sector. Resulting from this kind of exceptionally contagious virus and a shortfall of medical workers in the hospitals, front-line health workers, and patients are at risk. Thus, with an aim to diminish the risk of infections, a mobile robotic system is proposed that can autonomously ensure safety and protection in the hospital. The system can monitor the patients by moving autonomously and sanitizing the floor throughout the hospital, which is implemented by Robot Operating System (ROS), SLAM (Simultaneous Localization and Mapping) algorithm, and A∗ search algorithm, and then it uses the MobileNetV2 algorithm for safety mask detection and giving voice alert. The system also offers AI voice communication to assist and diagnose the patients, which can lessen person-to-person contact. The system has anticipated 89% accuracy for AI custom dataset, whereas the validation accuracy for face mask detection is 95%. © 2022 IEEE.

HYBRID AUDIT AS AN INNOVATIVE APPROACH IN THE AUDIT PRACTICE OF FORESTRY

Auditing has undergone many changes in the context of globalization and digitalization. Under the influence of the COVID-19 health crisis, additional restrictions are introduced, which are related to the impossibility of physical inspections (desk cheks or on-site ones). This has a significant effect on the activities of sectors and enterprises where on-site inspections are a priority, like in forestry. In order to answer the current challenges, the Forest StewardshipCouncil introduces a new audit approach, called "hybrid audit". The main goal of the study is to analyze the benefits and risks of hybrid audits in forestry. This will be done by (1) performing a comparative analysis with other types of audit (particularly financial and IT audit) and (2) deriving the characteristics and requirements of the hybrid audit. The study is conducted based on the scientific methods of analogy, analysis and synthesis, induction, deduction, and logical approach. The results of the study outlined the need for specific, local forestry-related rules and methodology of the procedures for the hybrid audit to be developed. Additionally, there is a risk of gaps in the check and verification of the forestry documentation and resources when using the hybrid audit. © 2022 15th International Scientific Conference WoodEMA 2022 - Crisis Management and Safety Foresight in Forest-Based Sector and SMES Operating in the Global Environment. All rights reserved.

Halliburton Saudi Arabia Crisis Management and Lessons Learn During COVID-19 Pandemic in the Kingdom of Saudi Arabia

During the first quarter of 2020, the world encountered a crucial and unprecedented health crisis. The global transmission of COVID-19 poses a significant challenging situation for Oil and Gas industry, particularly in the absence of standardized procedures and recognized methods. Like many other countries worldwide, Saudi Arabia implemented the lockdown for utmost public and private services and controlled population movement through curfew. With the execution of these tight mitigation requirements, Halliburton Saudi Arabia has been able to maintain business continuity by looking at the basic approach of health, safety, and environmental (HSE) processes through crisis management decision making and utilizing digital solutions. The purpose of this paper is to showcase how Halliburton Saudi Arabia developed sustainable adjustable process and methods that reduced exposure and the pandemic-related potential risks associated with working in offices, rig sites, workshops, and laboratories while maintaining business continuity in operation, manufacturing, and technology. Halliburton Saudi Arabia preformed risk analysis, tracking systems, exposure modification methodologies, communication strategies and management decisions that helped the company overcome challenges during the pandemic. The implementation of risk assessments, adaptable safety procedures and utilizing more than 5 digital platforms, served Halliburton employees and its work force throughout 2021 and into 2032. In this paper, we share lessons learned during the pandemic, how we overcame the unprecedented health crisis and how we continue to deal with the pandemic impact. Copyright © 2022, International Petroleum Technology Conference.

Hazmat Cooling Vest Temperature Using Heat Pipe and Thermoelectric

Corona Virus (COVID-19) is a virus that is endemic almost all over the world, including Indonesia. COVID-19 was first confirmed by the World Health Organization (WHO) on December 31, 2019, in Wuhan City, Hubei Province, China, and then rapidly expanded outside of China. To suppress the Covid-19 case, medical volunteers are needed as the main actors in efforts to handle Covid-19 patients. This makes health care facilities also need to focus on the principles of health worker safety, not only focus on the principles of patient safety. This also makes health care facilities also need to focus on the principles of health worker safety, not only focus on the principles of patient safety. The use of hazmat clothes is one of the efforts to protect health workers when in contact with Covid-19 patients. Hazmat clothes are technically referred to as "encapsulated waterproof protective clothing” which is PPE that must be used for officers from the risk of contracting the Covid-19 virus through airborne droplets and contact with patients and patient body fluids. Although hazmat clothing is an important PPE for health workers to stay protected, the use of hazmat clothing for a long time often makes medical personnel feel uncomfortable when providing services. Based on the problems above, the researchers conducted a study on the heat pipe - thermoelectric hazmat suit cooling vest. This technology can absorb more heat than other methods by simply applying the principle of capillarity to the wicks on the pipe walls. schematic of testing a cooling vest on a hazmat suit. The loading on the thermoelectric is given through the DC - Power supply. The temperature data read by the sensor will be detected by the computer system using the NI 9123 and C-DAQ 9174 modules. The test results can be viewed using the NI LabView 2017 software. The temperature used in this experiment is the result of tests carried out for 30 min. Based on the tests that have been carried out, the heat pipe-based thermoelectric hazmat suit cooling vest has been able to reach the lowest thermoelectric temperature of 24,42 ∘C, which is distributed through heat pipes to body parts. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Dynamic and Fuzzy Multi-Criteria Decision-Making for Regional Safety Assessment Ranking

This paper proposes augmenting the Multi-Criteria Decision-Making (MCDM) hybrid methodology of AHP-TOPSIS with dynamic-case handling (DCH) calculations and fuzzy logic. This method is evaluated with an illustrative example of three interrelated scenarios that rank 20 countries based on regional safety assessment related to the COVID-19 pandemic. The proposed method is compared to related work in the field. Additionally, sensitivity analysis is performed to evaluate the robustness of the proposed methodology. Empirical results demonstrate that the AHP-TOPSIS method coupled with fuzzy logic and DCH calculations is a realistic decision-making approach. © 2022 IEEE.

Quality control strategy of remote heart rate monitoring devices in a health center in Peru

Cardiovascular disease (CVD) is a broad term used to refer to heart diseases and blood vessels. According to the World Health Organization (WHO), the number of deaths from heart diseases between the years 2000-2019 has ranged between 2 million and almost 9 million people. After analyzing the need to implement and validate a monitoring system for the PUCP health center, we chose the FL10 device. This device has passed the design specifications under the following standards: IEC 60601-1:2005, IEC 60601-1-2:2014 and ISO 60601-2-47:2012, therefore it meets the safety, sensitivity and efficacy parameters but its design does not favor maintenance and periodic quality evaluations with commercial measurement equipment. An adapter was made by implementing digital manufacturing by 3D design using Inventor professional 2021 software with the educational license granted by the university. The electrical safety test showed that the leakage current value of all the electrodes was 0.3 uA which is less than the standard (0.1 mA) and the efficacy test showed that all measurements were acceptable. In the case of reproducibility and repeatability tests, the measurements were acceptable as well;in the value of 180 bpm, we observe an attenuation (179 bpm), however it is within the margin of error of 1 bpm. In summary, the FL 10 device is electrical safety and efficient in a health center in Peru. © 2022 IEEE.

Process Intensification and Increased Safety for the On-Demand Continuous Flow Synthesis of Dithiothreitol, a Crucial Component in Polymerase Chain Reaction Testing Kits

The importance of rapid access to diagnostics tools in the identification of pathogens-including their crucial component, bioreagents-was recently underscored in the COVID-19 pandemic. The currently adopted synthesis of dithiothreitol (DTT) involves four steps in batch with long reaction times and which generates a highly carcinogenic and mutagenic bis-epoxide intermediate. In this work, we have developed an intensified telescoped three-step continuous flow synthesis of DTT involving a base-mediated ring closure epoxidation, a nucleophilic epoxide opening with thioacetic acid, and an acid-mediated deacetylation. One of the key features is that the first two steps are conducted in a telescoped continuous flow fashion, allowing generation and consumption of the hazardous intermediate in situ, suppressing the need for its isolation, and improving the overall safety of the synthesis. The process is completed by an acid-catalyzed deacetylation and a subsequent recrystallization to afford the desired DTT. Flow chemistry allows here to intensify the process by using high temperatures and high pressures while minimizing the number of unit operations and improving the overall safety of the process. Our protocol permits the on-demand production of DTT in case of future outbreaks. © 2023 American Chemical Society.

Risk-Informed Decision-Making Tool for Covid-19 Community Behavior and Intervention Scenario Assessment

The spread of the COVID-19 pandemic across the world has presented a unique problem to researchers and policymakers alike. In addition to uncertainty around the nature of the virus itself, the impact of rapidly changing policy decisions on the spread of the virus has been difficult to predict. Using an epidemiological Susceptible-Infected-Recovered-Dead (SIRD) model as a basis, this paper presents a methodology for modeling many uncertain factors impacting disease spread, ultimately to understand how a policy decision may impact the community long term. The COVID-19 Decision Support (CoviDeS) tool, utilizes an agent-based time simulation model that uses Bayesian networks to determine state changes of each individual. The model has a level of interpretability more extensive than many existing models, allowing for insights to be drawn regarding the relationships between various inputs and the transmission of the disease. Test cases are presented for different scenarios that demonstrate relative changes in transmission resulting from different policy decisions. Further, we will demonstrate the model's ability to support decisions for a smaller sub-community that is contained in a larger population center (e.g. a university within a city). Results of simulations for the city of Los Angeles are presented to demonstrate the use of the model for parametric analysis that could give insight to other real-world scenarios of interest. Though improvements can be made in the model's accuracy relative to real case data, the methods presented offer value for future use either as a predictive tool or as a decision-making tool for COVID-19 or future pandemic scenarios. © 2022 Probabilistic Safety Assessment and Management, PSAM 2022. All rights reserved.

COVID-19 pandemic: Analyzing of infection waves in Germany in comparison with common infectious diseases by using reliability methods

Since December 2019, the world is confronted with the COVID-19 pandemic, caused by the Coronavirus SARS-CoV-2. The COVID-19 pandemic with its incredible spreading speed shows the vulnerability of a globalized and networked world. The first two years of the pandemic were characterized by several infection waves, described by length, peak, and speed. The infection waves caused a heavy burden on health systems and severe restrictions on public life, like educational system shutdown, travel restrictions, limitations regarding public life, or a comprehensive lockdown within a lot of countries. The goal of the presented research study is the analysis of the development of the six dominant infection waves in Germany within the first two years of the COVID-19 pandemic (February 2020 - February 2022). The analyses are focusing on the occurrence of infection and spreading behavior, in detail on attributes like length, peak, and speed of each wave. Furthermore, various impacts of lockdown strategies (hard, soft) or virus variants are considered. The analyses of the infection waves are based on a transfer and application of methods - especially the Weibull distribution model and statistical hypothesis tests - used in reliability engineering for analyzing the upcoming failure development within product fleets in the field. The spreading behavior of a COVID-19 infection wave can be described by the Weibull distribution model in a sound way, related to a short time interval. The interpretation of the Weibull model parameters allows the assessment of the COVID-19 infection wave characteristics and generates additional information to classical infection analysis models like the SIR model [10]. Finally, the characteristics of the COVID-19 infection waves are analyzed in the context of other common infectious diseases in Germany like Influenza or Norovirus. This study continues previous research;cf. [1-3,11,12]. © 2022 Probabilistic Safety Assessment and Management, PSAM 2022. All rights reserved.

Adaptation of a Study Design to the COVID-19 Pandemic Regulations - Evaluation of a Voice-Controlled Robotic Kitchen Assistance System for the Support of Older Adults in Need of Care

Background. The COVID-19 pandemic had and still has a major impact on the design and realization of studies with volunteers due to the associated restrictions in face-to-face research. In connection with vulnerable target groups in particular, alternative study designs have to be considered. In the AuRorA project, a voice-controlled robotic kitchen assistance system was developed for supporting older adults in need of care. We will present how we planned and conducted the evaluation of the system despite pandemic restrictions. Methods. We carried out an iterative risk management based on ISO 12100:2010 with a multidisciplinary team before starting the study. In addition to the consideration of personnel safety in human-robot interaction, additional focus was placed on subject safety with respect to the COVID-19 pandemic. Results. The risk management revealed that the study cannot be conducted via face-to-face with the target group. As a result, the study was performed via an online survey. We evaluated the robotic system using questionnaires. We simulated and filmed an exemplary cooking process and implemented the video into the survey. The target group was reached via mail. Discussion. Other research groups also struggeled to adapt their studies to the pandemic situation. However, no best practice example exist yet. With our risk management, we were able to adapt our study design. However, only part of the target group could be reached. In addition, it can be questioned wether the ISO, which adresses machine safety, was an efficient application of risk management in this area. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Managing 4 (Four) Major Offshore Projects Amid COVID 19 Pandemic - A Case Study from Health & Safety (H&S) and Quarantine Management

For an upstream oil and gas company, avoiding an offshore COVID-19 outbreak while executing four different offshore projects poses a huge challenge, particularly in a country experiencing a daily COVID-19 test positivity rate over 20%. Even minor mismanagement of the quarantine process can lead to an offshore COVID-19 outbreak, with the risk of shutting down campaigns and severely impacting business objectives. The challenge is therefore to avoid an offshore COVID-19 outbreak, ensuring well-being of personnel during the quarantine period and managing quarantine related costs, including COVID-19 test costs. To ensure effective quarantine management, a new approach was created that applied a combination of medical assessments, Health & Safety (H&S) and security measures. Quarantine management was led by a special task force responsible for ensuring the readiness of transportations, rooms, PCR tests, as well as overall compliance to quarantine rules. In compliance with government regulations and WHO recommendations, another complimentary approach was applied that sequestered personnel who tested positive in an isolation room. Effective quarantine management was established with the assistance of the company Business Continuity Management Team (BCMT). The company was able to complete four different major offshore projects with no offshore COVID-19 outbreaks. During these operations, over 1,000 personnel were quarantined and tested with a 5.37% positivity rate at the pre-work quarantine site. Confirmed cases were managed in full compliance with government regulations. The result of this effective quarantine management system, has allowed the company to achieve scorecard performance goals while delivering all four of the major offshore work-scopes, as per the original business plan. This paper discusses quarantine management as part of business continuity management covering medical assessment, H&S and security measures amidst a national COVID-19 pandemic. These programs were applied in an adaptive method-based risk assessment, which based on evidence base approaches, during frequently changing government regulations. Copyright © 2022, Society of Petroleum Engineers.

Augmented Reality Delivers Differential Value in Safety Assurance on Rigs Onshore Abu Dhabi During Covid-19 Pandemic Courtesy of the Wearable Camera

The COVID-19 pandemic posed a real challenge to business continuity of our safety assurance programs, incident investigations and training of young engineers which are traditionally conducted through physical rig site visits. The measures taken to control the spread of the COVID-19 virus meant that physical visits to the rig sites were restricted and during certain times completely stopped. The consequence of such measure meant that the safety assurance program was heavily impacted as the weekly leadership site visits were suspended. This posed a risk of a potential increase in safety incidents. Moreover, only safety critical personnel were allowed at the rig sites meaning that all trainee drilling supervisors and engineers were not allowed to visit the rig sites as part of the COVID-19 control measures. This in turn directly impacted the training program of our young engineers and trainee drilling supervisors for the entire COVID-19 pandemic period - Not Good! The potential risk of an increase in safety incidents due to lack of safety assurance was too great a cost to accept. Therefore, the drilling team started to think differently. A multi-disciplinary brainstorming workshop involving technology providers in the digital space was conducted to discuss possible ways of conducting rig site visit virtually. During this workshop, augmented reality via use of the wearable camera was flagged as a technology ready solution for our challenge. Several subsequent sessions were conducted with ADNOC Onshore IT team to build a business case for the wearable camera. The wearable camera is an intrinsically safe Zone 1 certified fully rugged head-mounted device made for harsh environments. The high-resolution micro display fits just below the user's line of sight and views like a 7" tablet. A business case was put forward and approved to conduct a proof of concept (PoC) to specifically test the ability to conduct remote rig visits, investigations and training programs. The PoC was successfully completed and allowed the team to conduct remote rig visits, audits, investigations and training programs with the "touch and feel" of being at the rig site - a paradigm shift in the way we run our business. Copyright © 2022, Society of Petroleum Engineers.

Compositional Simulation of Abstract State Machines for Safety Critical Systems

State Machines (ASMs) communicating through I/O events. The proposed method allows the co-simulation of ASM models of separate subsystems of a Discrete Event System in a straight-through processing manner according to a predefined orchestration schema. We also present our experience in applying and validating the proposed technique in the context of the MVM (Mechanical Ventilator Milano) system, a mechanical lung ventilator that has been designed, successfully certified, and deployed during the COVID-19 pandemic. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Obsolescence Assessment Approach: Case of Mechanical Ventilators Under the Covid-19 Environment

Support treatment for patients with severe Covid19 symptoms is based on mechanical ventilation. As a result of the spread of the coronavirus, various medical institutions found it necessary to acquire more mechanical ventilators, causing an increase in the demand for their production. The production, distribution, and use conditions present a potential medium- and long-term risk concerning the equipment’s operation, safety, and effectiveness. In the face of the pandemic, the care that must be implemented to guarantee patient safety, the effectiveness of the equipment, and the quality of care, should be a high priority. This paper presents a first approach to analyzing the causes that can lead to a potential state of obsolescence in mechanical ventilators used to care for patients with Covid19. We collected information about mechanical ventilators, identified the factors that may cause a failure in the device, applied Multi-Criteria Decision Analysis for three possible outcomes, and compared the results with the expert opinion of clinical engineers. We found a total of 30 factors and classified them into associated, operation, and external concerning the ventilator. The results show that the factors found are associated with the technical field. The knowledge of the operation of the equipment’s subsystems and familiarity with its components is essential. The contrast with the expert opinion and the non-inclusion of factors such as the economic ones led to the development of more in-depth work on this topic. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

An Optimized and Safety-aware Maintenance Framework: A Case Study on Aircraft Engine

The COVID-19 pandemic has recently exacerbated the fierce competition in the transportation businesses. The airline industry took one of the biggest hits as the closure of international borders forced aircraft operators to suspend their international routes, keeping aircraft on the ground without generating revenues while at the same time still requiring adequate maintenance. To maintain their operational sustainability, finding a good balance between cost reductions measure and safety standards fulfillment, including its maintenance procedure, becomes critical. This paper proposes an AI-assisted predictive maintenance scheme that synthesizes prognostics modeling and simulation-based optimization to help airlines decide their optimal engine maintenance approach. The proposed method enables airlines to utilize their diagnostics measurements and operational settings to design a more customized maintenance strategy that takes engine operations conditions into account. Our numerical experiments on the proposed approach resulted in significant cost savings without compromising the safety standards. The experiments also show that maintenance strategies tailored to the failure mode and operational settings (that our framework enables) yield 13% more cost savings than generic optimal maintenance strategies. The generality of our proposed framework allows the extension to other intelligent, safety-critical transportation systems. © 2022 IEEE.