Imperfect Monitoring Research Articles

Maintenance Planning under Imperfect Monitoring: an Efficient POMDP Model Using Interpolated Value Function

We develop in this paper a partially observable Markov decision process (POMDP) model for a maintenance planning problem and solve it with an efficient point-based value iteration (PBVI) algorithm. We consider a single-unit system, subjected to random degradation and failures, and for which the current degradation state can be partially observed via an imperfect monitoring system. The system state space is finite, and we model the following maintenance operations: i) perfect inspection, ii) preventive maintenance and iii) corrective maintenance. The goal is to optimize the maintenance policy by taking into account the imperfect monitoring data in order to minimize the expected discounted maintenance cost over an infinite time horizon. We formulate the problem as a POMDP where, at each time step, it should be decided whether or not to conduct a maintenance operation, and if so, which one. To keep the model general and flexible, we suppose that monitoring data are collected every K time steps (i.e. one observation epoch). The model is completed by a constraint imposing that only one maintenance operation can be conducted per observation epoch. Eventually, we solve it using a PBVI algorithm. The value function is approximated by interpolation of grid data points, and new relevant points are dynamically added into the grid where they most improve the value function. This approach is compared to a POMDP modeling based on approximate sample paths (ASP); when evaluated in different cost scenarios, the proposed approach systematically finds better maintenance policies for a comparable computation time. The computation of a lower bound finally proves that we are able to get the optimal value of the problem with satisfying precision.

Leadership with Imperfect Monitoring

This paper provides experimental evidence on how monitoring intensity shapes the impact of leadership instruments like leading-by-example and punishment. The results show that, with low monitoring intensity, neither leading-by-example nor punishment increases cooperation if the leader can already send nonbinding signals about desired behavior. We identify two different reasons for this effect. Regarding leading-by-example, it is the cautiousness of the leader. Leaders are reluctant to recommend cooperative behavior and act accordingly, even though followers are particularly reciprocal in this context. Regarding punishment, it is the level of monitoring that matters. Monitoring of individual follower behavior increases the cooperation of leaders and followers across all treatments, but in particular, if the leader can punish uncooperative behavior. This result implies that monitoring in itself does not have a negative impact on the inclination to cooperate. It suggests that any motivational crowding out effect derives from a leader’s choice of monitoring, as it signals low trust in the followers. The paper concludes with a discussion of the implications.

Procedural Guide for System-Level Impact Evaluation of Industrial Artificial Intelligence-Driven Technologies: Application to Risk-Based Investment Analysis for Condition Monitoring Systems in Manufacturing.

Industrial artificial intelligence (IAI) and other analysis tools with obfuscated internal processes are growing in capability and ubiquity within industrial settings. Decision-makers share their concern regarding the objective evaluation of such tools and their impacts at the system level, facility level, and beyond. One application where this style of tool is making a significant impact is in Condition Monitoring Systems (CMSs). This paper addresses the need to evaluate CMSs, a collection of software and devices that alert users to changing conditions within assets or systems of a facility. The presented evaluation procedure uses CMSs as a case study for a broader philosophy evaluating the impacts of IAI tools. CMSs can provide value to a system by forewarning faults, defects, or other unwanted events. However, evaluating CMS value through scenarios that did not occur is rarely easy or intuitive. Further complicating this evaluation are the ongoing investment costs and risks posed by the CMS from imperfect monitoring. To overcome this, an industrial facility needs to regularly and objectively review CMS impacts to justify investments and maintain competitive advantage. This paper's procedure assesses the suitability of a CMS for a system in terms of risk and investment analysis. This risk-based approach uses the changes in the likelihood of good and bad events to quantify CMS value without making any one-time point-wise estimates. Fictional case studies presented in this paper illustrate the procedure and demonstrate its usefulness and validity.

Abundance decline in the avifauna of the European Union reveals cross-continental similarities in biodiversity change.

Although global assessments provide evidence of biodiversity decline, some have questioned the strength of the evidence, with local assemblage studies often showing a more balanced picture of biodiversity change. The multifaceted nature of biodiversity and imperfect monitoring datasets may partially explain these findings. Here, using an extensive dataset, we find significant biodiversity loss in the native avifauna of the European Union (EU). We estimate a decline of 17–19% in the overall breeding bird abundance since 1980: a loss of 560–620 million individual birds. Both total and proportional declines in bird numbers are high among species associated with agricultural land. The distribution of species’ population growth rates (ln) is centered close to zero, with numerical decline driven by substantial losses in abundant species. Our work supports previous assessments indicating substantial recent biodiversity loss and calls to reduce the threat of extinctions and restore species’ abundances, for the sake of nature and people.

When a Generalized Linear Model Meets Bayesian Maximum Entropy: A Novel Spatiotemporal Ground-Level Ozone Concentration Retrieval Method

In China, ground-level ozone has shown an increasing trend and has become a serious ambient pollutant. An accurate spatiotemporal distribution of ground-level ozone concentrations (GOCs) is urgently needed. Generalized linear models (GLMs) and Bayesian maximum entropy (BME) models are practical for predicting GOCs. However, GLMs have limited capacity to capture temporal variations and can miss some short-term and regional patterns, while the performance of BME models may degrade in cases of sparse or imperfect monitoring networks. Thus, to predict nationwide 1 km monthly average GOCs for China, we designed a novel hybrid model containing three modules. (1) A GLM was established to accurately describe the variability in GOCs in the space domain. (2) A BME model incorporating GLM residuals was employed to capture the temporal variability of GOCs in detail. (3) A combination of GLM and BME models was developed based on the specific broad range of each submodel. According to the cross-validation results, the hybrid model exhibited superior performance, with coefficient of determination (R2) values of 0.67. The predictive performance of the large-scale and high-resolution hybrid model is superior to that in previous studies. The nationwide spatiotemporal variability of the GOCs derived from the hybrid model shows that they are valuable indicators for ground-level ozone pollution control and prevention in China.

Supply‐ vs. Demand‐Side Transparency: The Collusive Effects Under Imperfect Public Monitoring*

We analyse how market transparency affects collusion under imperfect monitoring where punishment phases occur on the equilibrium path. We show that increased transparency causes a ‘pro‐competitive’ demand‐side effect and an ‘anti‐competitive’ supply‐side effect on the optimal symmetric perfect public equilibrium (SPPE) profits. When transparency increases on both sides of the market, the optimal SPPE profits unambiguously increase at the perfect monitoring limit, because the pro‐competitive demand‐side effect vanishes. This result holds even when there is minimal structure on the competition game. The supply‐side effect also dominates away from the limit under reasonable conditions. We draw conclusions for policy.

The Effect of Tightening Standards on Automakers’ Non‐compliance

This study investigates how tightening standards can result in greater non‐compliance, especially when market and regulatory interests are misaligned. We confirm a causal relationship that explains the highly publicized auto industry non‐compliance phenomenon where on‐road NOx emissions exceeded standards. Based on a 15‐year on‐road vehicle emissions dataset covering 148,837 vehicles from 42 automakers in the EU, we use regression discontinuity to identify the causal impact of standards tightening on non‐compliance by controlling other confounding factors. Our results suggest that in the absence of effective monitoring, tightening standards directly drives up automakers’ non‐compliance. Furthermore, we find that automakers facing more intense substitution pressure from competitors or with less advanced emissions control technology have a higher non‐compliance rate. Our findings speak to both policymakers as well as managers in the private sector. When setting limit‐based performance goals in situations with conflicting interests and imperfect monitoring, they should anticipate non‐compliance from the regulated parties. Our results suggest that tightening standards in such situations should be accompanied by stricter monitoring or other actions that discourage non‐compliance.

Resilient monitoring in self-adaptive systems through behavioral parameter estimation

Cyber-physical systems need self-adaptation as a mean to autonomously deal with changes. For runtime adaptation, a cyber-physical system repeatedly monitors the environment for detecting possible changes. Faults in the monitoring devices due to the dynamic and uncertain environment is very likely, necessitating resilient monitoring. In this paper, we discuss imperfect monitoring in self-adaptive systems, and propose a model-driven methodology to represent the self-adaptive system using a parametric Markov decision process, where the changes are reflected by a set of model parameters. Fault in the monitoring device may result in some parameter valuation miss. We propose a comprehensive framework for parameter estimation using behavioral patterns of the system implemented by a pattern-matching component. The proposed method simulates the current behavior of the system using random walk patterns, and matches it to a history of patterns for estimating the valuation of the omitted data. The results show that the accuracy of the framework is evaluated about 96% under imperfect monitoring. In addition, we elaborate a set of theoretical proofs to support rigorous error analysis, and determine a certain upper-bound of errors to guarantee an accurate decision-making outcome. We establish a logical connection between the error and the accuracy of decisions, and introduce tolerable error metric that is used to guarantee the accuracy of decisions under estimation.

Collective Incentives and Cooperation with Imperfect Monitoring

We experimentally explore the role of collective incentives in sustaining cooperation in finitely repeated public goods games with imperfect monitoring. In our experiment players only observe noisy signals about individual contributions, while total output is perfectly observed. We consider sanctioning mechanisms that allow agents to commit to collective punishment in case total output falls short of a target. We fi nd that cooperation is higher in the case of collective punishment compared to both the case of no punishment and the case of standard peer-to-peer punishment which conditions on the noisy signals. Further experiments indicate that both the commitment possibility and the collective nature of punishment matter for the positive effect of collective incentives on cooperation.

Algorithmic collusion with imperfect monitoring

We show that if they are allowed enough time to complete the learning, Q-learning algorithms can learn to collude in an environment with imperfect monitoring adapted from Green and Porter (1984), without having been instructed to do so, and without communicating with one another. Collusion is sustained by punishments that take the form of “price wars” triggered by the observation of low prices. The punishments have a finite duration, being harsher initially and then gradually fading away. Such punishments are triggered both by deviations and by adverse demand shocks.

Field experiment using transient energy method to locate a single-phase to ground fault

Distribution networks in China and several other countries are predominantly neutral inefficiently grounding systems (NIGSs), and more than 80% of the faults in distribution networks are single-phase-to-ground (SPG) faults. Because of the weak fault current and imperfect monitoring equipment configurations, methods used to determine the faulty line sections with SPG faults in NIGSs are ineffective. The development and application of distribution-level phasor measurement units (PMUs) provide further comprehensive fault information for fault diagnosis in a distribution network. When an SPG fault occurs, the transient energy of the faulted line section tends to be higher than the sum of the transient energies of other line sections. In this regard, transient energy-based fault location algorithms appear to be a promising resolution. In this study, a field test plan was designed and implemented for a 10 kV distribution network. The test results demonstrate the effectiveness of the transient energy-based SPG location method in practical distribution networks.

Imperfect monitoring and informal insurance: The role of social ties

This paper examines whether social ties impact informal insurance in the presence of asymmetric information. Using games with residents of informal settlements in Kenya, I vary the observability of task completion and partners. While individuals are less likely to make transfers as a result of imperfect monitoring, socially close individuals are 30% more likely to make transfers. As a result, socially close individuals are 47% more likely to make transfers than socially distant individuals when task completion cannot be observed. I then explore the mechanisms by which social connections influence risk sharing. I find evidence that social connections increase participants’ motivation to work only when task completion cannot be observed, potentially due to social collateral and/or intrinsic motivation.

Artificial Intelligence: Can Seemingly Collusive Outcomes Be Avoided?

Strategic decisions are increasingly delegated to algorithms. We extend the results of Waltman and Kaymak [2008] and Calvano et al. [2020b] to the context of dynamic optimization with imperfect monitoring by analyzing a setting where a limited number of agents use simple and independent machine-learning algorithms to buy and sell a storable good on behalf of a large number of consumers. No specific instruction is given to them, only that their objective is to maximize profits based solely on past market prices and payoffs. With an original application to battery operations, we observe that the algorithms learn quickly to exert market power at seemingly collusive levels, despite the absence of any formal communication between them. Contrary to the findings reported in the existing literature, we show that seeming collusion may originate in imperfect exploration, rather than excessive algorithmic sophistication. We then show that a regulator may succeed in disciplining the market to produce socially desirable outcomes by enforcing decentralized learning or with adequate intervention during the learning process.

Endogenous Monitoring in a Partnership Game

I consider a repeated game in which, due to imperfect monitoring, no collusion can be sustained. I add a self-interested monitor who commits to obtain private signals of firms’ actions and sends a public message. The monitor makes an offer specifying the precision of the signals obtained and the amount to be paid in return. First, with a low monitoring cost, collusive equilibria exist. Second, collusive equilibria are monitor-preferred. Third, in monitor-preferred equilibria, firms’ payoffs are decreasing in the discount factor. My model helps explain cartel agreements between self-interested parties and firms in legal industries in the United States and Europe. (JEL C73, D43, D82, L12)

Repeated partnerships with multiple equilibria and imperfect monitoring: An experimental study

I investigate finitely repeated partnership games with imperfect monitoring where both mutual effort and mutual shirking are Nash equilibria of the stage game. The treatment variable is the number of repetitions. I find that period 1 effort rates are increasing in the number of repetitions, but subjects use trigger strategies that switch to permanent shirking after enough failed projects so effort rates decrease across time in all treatments. Additionally, the rate of decrease is less when there are more repetitions. Total effort does not vary much across the treatments, because the increased period 1 effort when the number of repetitions is large is mostly canceled out by the erosion of effort as the game progresses.

Deferred Pay in Financial Services: Compliance, Productivity and Attracting Talent

Behavior that violates company policies and/or societal notions of ethical conduct can sometimes produce short-term profits and consequent bonuses. Due to imperfect monitoring, such behavior may not be identified until after bonuses have been awarded and paid. Deferred remuneration with malus provisions has been proposed as a possible antidote to misconduct in the financial services industry. The principal attraction of deferred remuneration is the potential for better monitoring of behavior and outcomes prior to the payment of any variable remuneration. This paper is the first to empirically examine the impact of deferred remuneration on both compliance and productive behavior. We also consider how this change might affect the ability of the industry to attract talent through investigation of self-selection effects. With 298 student participants, we first use experimental methods to examine the case where a change in remuneration is imposed. We observe an increase in strategic violations of policy when deferred payment of variable remuneration is imposed, and mixed results for compliance behavior depending on how it is measured. Allowing for self-selection effects, where individuals can choose their preferred payment structure, the benefits of deferrals become more apparent. Relative to the case of immediate payment, deferred payment of variable remunerations improves compliance on all our measures. For example, the proportion of participants with zero policy violations is 23.2% higher and we also observe a statistically significant increase in productivity. We show that more productive individuals are attracted to the condition with deferred payment and superior monitoring, while less compliant individuals are attracted to the condition with immediate payment and poor monitoring. Finally, we observed that male participants in our experiment are more likely than females to select the condition with immediate payment and poor monitoring. The study suggests that the short-term impact of a switch to deferred remuneration would be limited. Over the long-term, allowing for self-selection effects to occur, adoption of deferred remuneration is likely to produce improved conduct, higher productivity and a greater proportion of females in the workforce, relative to workplaces retaining immediate payment of variable remuneration. A switch to fixed remuneration, with neither sanctions nor benefit flowing from policy violations, would not be beneficial for employee behavior. After allowing for self-selection effects, this study confirms that productivity decreases under fixed remuneration and suggests that there would be no compliance benefits.

Extortion and Cooperation in Rating Protocol Design for Competitive Crowdsourcing

Although crowdsourcing has emerged as a paradigm for leveraging human intelligence and activity to solve a wide range of tasks, strategic workers will find enticement in their self-interest to free-ride and attack in a crowdsourcing contest dilemma game. Existing incentive mechanisms are not effective to avoid socially undesirable equilibrium due to the following features of competitive crowdsourcing: in the presence of imperfect monitoring, heterogeneous workers with competing interest tend to beat their opponents for larger self-profit, and the fact that they can freely and frequently change their opponents makes the situation much more complicated. Taking these features into consideration, this article proposes a mechanism design problem to enforce cooperation and extort selfish works simultaneously, with the objective of maximizing the requester’s utility. To solve the problem, we integrate binary ratings with differential pricing to develop a novel rating protocol. By establishing a mathematical model for the problem and quantifying necessary and sufficient conditions for a sustainable social norm, we provide design guidelines for optimal rating protocols and design a low-complexity algorithm to select optimal design parameters. Finally, extensive evaluation results demonstrate the performance of our proposed rating protocol and reveal how intrinsic parameters impact on design parameters.

Monitoring and punishment networks in an experimental common pool resource dilemma

AbstractWith the aid of a lab experiment, we explored how imperfect monitoring and punishment networks impacted appropriation, punishment and beliefs in a common pool resource appropriation dilemma. We studied the differences between the complete network (with perfect monitoring and punishment, in which everyone can observe and punish everyone else) and two ‘imperfect’ networks (that systematically reduce the number of subjects who could monitor and punish others): the directed and undirected circle networks. We found that free riders were punished in all treatments, but the network topology impacted the type of punishment: the undirected circle induced more severe punishment and prosocial punishment compared to the other two networks. Both imperfect networks were more efficient because the larger punishment capacity available in the complete network elicited higher punishment amount.

The Optimality of Ad Valorem Contracts

We provide a new theory of ad valorem contracts (i.e., contracts that vary with the value of the transaction), which can explain why such contracts are widely used between vertically related parties (e.g., in franchising and licensing). Ad valorem contracts allow upstream firms (principals) to preserve their own incentives to make ongoing investments in the channel and deal with pricing distortions caused by channel coordination problems, while at the same time adjusting their investment on the basis of demand shocks that are only observed by the downstream firms (agents). We show that the optimal ad valorem contract allows the principal to achieve the same profits as if it could observe the demand shocks and control price. This optimal contract makes use of revenue sharing (to balance investment incentives and make the principal’s investment responsive to demand through price), upfront fixed fees (to extract the agents’ expected profit), and an additional term that depends nonlinearly on either price or demand (to correct for remaining pricing distortions). Our results are robust to the introduction of competition between agents, production costs, and imperfect monitoring of the agents’ prices. This paper was accepted by Eric Anderson, marketing.

Defense against unknown broadband jammer for time‐critical operation in smart grid

In this work, the authors consider the communication network of a power substation, where multiple intelligent electronic devices (IEDs) transmit their delay sensitive control and monitoring messages to a common receiver over a wireless network; in the presence of a broadband jammer, which is capable of jamming multiple channels simultaneously. The objective of the IEDs is to successfully transmit their messages within a specified time, whereas the jammer wants to obstruct IED's transmission. A novel utility function is designed for the players, that addresses the time-critical nature of communication. Due to the conflicting interest of the IEDs and the jammer, they model the interaction between them as a repeated Bayesian zero-sum game, which also addresses the repeated interaction among the IEDs and the jammer, and the unavailability of exact information about the jammer. The equilibrium strategies for both the scenarios of perfect and imperfect monitoring are derived and verified through simulation results. Further, the performance of the proposed game model in various scenarios is thoroughly compared in the result section. Finally, the efficacy of the proposed defence strategies is tested in a practical communication network of a power substation under jamming, which is simulated in Optimised Network Engineering Tool (OPNET).