Industrial applications involving digital twins (e.g., behavior simulation) demand highly accurate, low-latency data, making real-time data acquisition critical. To meet performance demands, devices that do not support asynchronous communication need to acquire data at high frequency. In cloud-edge collaboration schemes, edge computing nodes typically acquire the data. However, high-frequency data acquisition and processing impose considerable costs, posing significant challenges for these resource-constrained nodes. To address this problem, we propose a model called Cost-minimized Data Edge Access (CDEA) that can dynamically minimize the edge costs while satisfying long-term performance requirements. CDEA quantifies data performance by decomposing the workflow of industrial systems into basic action units. These units are used to model data acquisition, data processing, data transmission, and cloud computing. Then, a cost minimization problem is formulated based on these components. To address irregular data changes and the general lack of available statistics on system’s network status, the framework incorporates Lyapunov optimization to transform the long-term guarantee over data performance into a series of instantaneous decision problems. Finally, a heuristic algorithm identifies the optimal data acquisition strategy. To validate CDEA’s effectiveness, we implemented it in two representative digital twin scenarios: cathode plate stripping and AGV transportation. Experimental results demonstrate that CDEA can indeed reduce both edge costs and cloud resources consumption while still ensuring high data performance.

Multi-agent path finding (MAPF) in dynamic and complex environments is a highly challenging task. Recent research has focused on the scalability of agent numbers or the complexity of the environment. Usually, they disregard the agents' physical constraints or use a differential-driven model. However, this approach fails to adequately capture the kinematic and dynamic constraints of real-world vehicles, particularly those equipped with Ackermann steering. This paper presents a novel algorithm named MARF that combines multi-agent reinforcement learning (MARL) with a Frenet lattice planner. The MARL foundation endows the algorithm with enhanced generalization capabilities while preserving computational efficiency. By incorporating Frenet lattice trajectories into the action space of the MARL framework, agents are capable of generating smooth and feasible trajectories that respect the kinematic and dynamic constraints. In addition, we adopt a centralized training and decentralized execution (CTDE) framework, where a network of shared value functions enables efficient cooperation among agents during decision-making. Simulation results and real-world experiments in different scenarios demonstrate that our method achieves superior performance in terms of success rate, average speed, extra distance of trajectory, and computing time.

Flexible manufacturing significantly enhances production adaptability but imposes higher requirements on the design efficiency of automation control systems. Mining existing project information and transforming it into reusable function blocks is an effective approach to achieving efficient development of industrial control programs. However, the precision of manually generated function blocks depends on the individual understanding of the developer, which lacks stability and can easily cause misunderstandings in later use and maintenance. Therefore, we proposed a graph-based intelligent wrapping framework for industrial control program snippets to enhance the normalization of custom function blocks. The framework utilizes service computing and LLMs to intelligently assist in generating function blocks. The whole process includes program graph representation, semantic understanding, and automatic organization. Finally, we evaluated the proposed framework through comparative experiments in three practical projects that involved up to$\mathbf{1 0 0}$function blocks.

Recent research shows that data augmentation can mitigate data sparsity and improve the robustness of sequential recommendation. However, many augmentation strategies operate on raw sequences directly, easily disrupting the inherent semantic and temporal organization of user behaviors. To address this issue, we propose a Semantic-Aware Sequence Augmentor for recommendation that disentangles user sequences into stable (long-term) and spontaneous (short-term) latent factors, then selectively augments only the short-term component. This design preserves users’ primary semantic context while injecting controlled diversity into their short-term interests. Underlying our approach is a diffusion-based procedure that generates coherent augmented sequences without compromising semantic integrity. Extensive experiments on six real-world datasets confirm the effectiveness of SASA, showing clear improvements over conventional augmentation methods.

This work focuses on multi-floor indoor exploration, which remains an open area of research. Compared to traditional methods, recent learning-based explorers have demonstrated significant potential due to their robust environmental learning and modeling capabilities, but most are restricted to 2D environments. In this paper, we proposed a learning-integrated topological explorer, LITE, for multi-floor indoor environments. LITE decomposes the environment into a floor-stair topology, enabling seamless integration of learning or non-learning-based 2D exploration methods for 3D exploration. As we incrementally build floor-stair topology in exploration using YOLO11-based instance segmentation model, the agent can transition between floors through a finite state machine. Additionally, we implement an attention-based 2D exploration policy that utilizes an attention mechanism to capture spatial dependencies between different regions, thereby determining the next global goal for more efficient exploration. Extensive comparison and ablation studies conducted on the HM3D and MP3D datasets demonstrate that our proposed 2D exploration policy significantly outperforms all baseline explorers in terms of exploration efficiency. Furthermore, experiments in several 3D multi-floor environments indicate that our framework is compatible with various 2D exploration methods, facilitating effective multi-floor indoor exploration. Finally, we validate our method in the real world with a quadruped robot, highlighting its strong generalization capabilities.

Image matching is an essential part of many processes in practical industrial applications. Using optimization algorithms to optimize the actual problems in industrial production can lead to more efficient use of resources. This paper presents a new algorithm called Beetle Swarm with Constrained L$\acute {e}$vy Flight (BSL) algorithm for solving problems in industrial production where image matching cannot be done quickly and accurately. This algorithm is based on the beetle antennae search algorithm. It combines the swarm intelligence algorithm with the constrained L$\acute {e}$vy flight and quickly finds the optimal solution through the long-horned beetle’s judgment of the left and right odour concentration, which reduces the blindness of L$\acute {e}$vy flight and dramatically improves the convergence speed of the algorithm. In the performance test, compared with other commonly used meta-heuristic algorithms, BSL shows stable performance with low time cost of convergence and better fitness results, which means BSL avoids the effect of random direction caused by a single beetle. In addition, this algorithm significantly improves the speed and accuracy of image matching in the application of Printed Circuit Board (PCB) defect detection. In practical application tests, the BSL algorithm was faster than the BAS algorithm, with a reduction in the fitness value. The subsequent robustness experiments further prove that the BSL algorithm has better noise immunity and is more suitable for application in actual production than Cuckoo Search (CS) and beetle antennae search (BAS).Note to Practitioners—This paper was motivated by the problem of matching images in PCB defect detection. Existing meta-heuristic approaches cannot achieve quick and accurate image matching due to the low convergence speed. This paper suggests a method that combines the beetle antennae search algorithm with the constrained L$\acute {e}$vy flight. In this paper, we improve the L$\acute {e}$vy flight mechanism to update the next position of each beetle to eliminate invalid solutions. Then, new directions are generated in combination with the odour concentration around the aspen whiskers, which reduces the blindness of L$\acute {e}$vy flight to a certain extent and, indeed, leads to the optimal solution. We also incorporate the Normalized cross-correlation (NCC) algorithm to improve the matching speed and noise immunity. The application in automated optical inspection (AOI) shows that the algorithm significantly improved the speed and accuracy when applied to production.

Accurate traffic flow forecasting is critical for the efficient operation of intelligent transportation systems (ITS), as it directly supports urban management and decision-making. With the increasing complexity of urban traffic, existing models often fail to fully capture the dynamic dependencies between traffic inflows and outflows. Some treat them as a unified process, while others only explore their commonalities. Inflows and outflows exhibit distinct patterns and interactions that require more refined modeling. To improve modeling performance, we propose BiSTGNN, a novel deep spatio-temporal network model which explicitly models bidirectional traffic flows as independent stochastic processes. Our approach leverages the unique temporal and spatial dependencies of each flow direction to distinguish transitions between directions, and integrates them through a composite graph convolution framework, offering a more detailed analysis of the transfer process between flows. Additionally, we introduce an innovative dynamic graph construction method that differentiates the interactions between inflows and outflows, capturing their heterogeneous relationships. Extensive experiments on five real-world traffic datasets demonstrate that our method outperforms state-of-the-art baselines, achieving superior accuracy in predicting both inflows and outflows.

LiDAR panoptic segmentation (LPS) performs semantic and instance segmentation for things (foreground objects) and stuff (background elements), essential for scene perception and remote sensing. While most existing methods separate these tasks using distinct branches (i.e., semantic and instance), recent approaches have unified LPS through a query-based paradigm. However, the distinct spatial distributions of foreground objects and background elements in large-scale outdoor scenes pose challenges. This paper presents DQFormer, a novel framework for unified LPS that employs a decoupled query workflow to adapt to the characteristics of things and stuff in outdoor scenes. It first utilizes a feature encoder to extract multi-scale voxel-wise, point-wise, and BEV features. Then, a decoupled query generator proposes informative queries by localizing things/stuff positions and fusing multi-level BEV embeddings. A query-oriented mask decoder uses masked cross-attention to decode segmentation masks, which are combined with query semantics to produce panoptic results. Extensive experiments on large-scale outdoor scenes, including the vehicular datasets nuScenes and SemanticKITTI, as well as the aerial point cloud dataset DALES, show that DQFormer outperforms superior methods by +1.8%, +0.9%, and +3.5% in panoptic quality (PQ), respectively. Code is available at https://github.com/yuyang-cloud/DQFormer.

Graph Convolutional Networks (GCNs) are currently the most widely used method for processing graph-structured data. However, recent research has revealed that the performance of GCNs dramatically decreases when confronted with adversarial attacks. This severely hinders their application in security-critical domains. Therefore, the development of GCNs that are resilient to adversarial attacks has emerged as a prominent research focus. Despite this, most current defense models with complex network architectures and optimization objectives are typically designed based on specific feature assumptions or attack manifestations, and do not enhance the inherent robustness of GCNs. They also overlook the changes induced by perturbations of varying intensities and the difference in attack phenomenon across different datasets. In response to this, we have delved into the impact of adversarial attacks on the spectrum, and propose an effective adaptive robust spectrum filter GCN (ASF-GCN). This approach enhances the robustness of GCN models through adaptive filtering without introducing additional conditional assumptions. We theoretically analyze that graphs have different robust frequency intervals under different conditions, validating the necessity of adaptive filtering. Additionally, we elucidate the role of degree distribution and maximum eigenvalue in adaptation. Extensive experiments on real-world graphs reveal that our model surpasses other defense models in overall performance.

The secure access of numerous heterogeneous devices ensures the stability of the industrial Internet of Things. Centralized authentication can be overwhelmed by an influx of authentication requests from malicious devices. Distributed device-to-device authentication is vulnerable to tampering attack evidence. Blockchain authentication eases evidence tampering, but traditional blockchains impose high-performance requirements on devices, rendering them unsuitable for resource-limited devices. Therefore, in this article, we apply IOTA to device access authentication. To the best of authors' knowledge, this is the first application. Based on this, we propose a blockchain-enabled distributed authentication mechanism for industrial device access. The authentication involves a consensus phase based on optimized IOTA and a unique identification code validation phase. We optimized the tip selection algorithm of IOTA to make it faster and more stable. The performance experiment results demonstrate that our mechanism meets the time-consuming requirements of device access. The security experiments indicate that the authentication phases effectively intercept the attacks.

The rise of smart factories is driving the complete automation of manufacturing environments, making anomaly detection a more important task than ever. With highly intelligent equipment deployed on fully-automated production lines, even a slight anomaly may seriously impact the entire manufacturing process. However, most industrial data containing slight anomalous features exhibits strong correlations that are not fully exploited by existing machine learning models. In general, the anomaly detection model needs to run within programmable logic controllers (PLCs) since it is the main controller of the intelligent equipment, and PLCs have limited resources making it difficult to execute larger models effectively. To address the challenge, we propose a graph anomaly detection method in programmable logic controllers based on service computing (PCSC). The model establishes a data relationship graph through clustering and then feeds it into a neural network with a symmetric structure consisting of graph convolutional layers and LSTM units. This approach enables the analysis of correlations within industrial data and facilitates the extraction of slight abnormal features. For efficient execution of the model, we establish service computing nodes in the PLCs that support model splitting. We tested the model on several publicly available datasets and an actual dataset from an injection molding production line. The results show that the model performs well on different datasets.

Multi-access Edge Computing (MEC) is an emerging architecture that extends cloud computing services to the edge of the network using mobile base stations. In a typical MEC paradigm edge nodes directly access heterogeneous nodes (e.g., sensors, devices, machines). However, limited to massive communication protocols of heterogeneous nodes, as well as the poor mobility, difficulty in lightweight, and low access efficiency brought by the current protocol parsing system based on the universal parser, create new challenges in the Internet of Things. We propose a dynamic protocol parsing system with optimized edge containers, called DPPS. It is suitable for MEC, allows edge nodes to dynamically access heterogeneous nodes, and supports cloud protocol parsing image management and edge node fault repair. Besides, we design a built-in container optimization algorithm based on the mutation cuckoo algorithm, which overcomes the memory resource shortage and energy consumption problems in DPPS. This algorithm demonstrates superior performance compared to other algorithms for embedded container optimization.

Siamese network has been a de facto benchmark framework for 3D LiDAR object tracking with a shared-parametric encoder extracting features from template and search region, respectively. This paradigm relies heavily on an additional matching network to model the cross-correlation/similarity of the template and search region. In this paper, we forsake the conventional Siamese paradigm and propose a novel single-branch framework, SyncTrack, synchronizing the feature extracting and matching to avoid forwarding encoder twice for template and search region as well as introducing extra parameters of matching network. The synchronization mechanism is based on the dynamic affinity of the Transformer, and an in-depth analysis of the relevance is provided theoretically. Moreover, based on the synchronization, we introduce a novel Attentive PointsSampling strategy into the Transformer layers (APST), replacing the random/Farthest Points Sampling (FPS) method with sampling under the supervision of attentive relations between the template and search region. It implies connecting point-wise sampling with the feature learning, beneficial to aggregating more distinctive and geometric features for tracking with sparse points. Extensive experiments on two benchmark datasets (KITTI and NuScenes) show that SyncTrack achieves state-of-the-art performance in realtime tracking.

In the era of the Industrial Internet of Things, anomaly detection is important for real-world applications. However, most streaming data lack meaningful labels. Furthermore, some anomalies of streaming data may be concept drift, but few methods can deal with it. To address these challenges, we propose an unsupervised anomaly detection solution that can deal with streaming data, called OOA-UADS (Offline, Online, Analysis-an Unsupervised Anomaly Detection Solution for Multivariate Time Series). The solution consists of three stages: offline training, online prediction and anomaly analysis. Time convolutional networks and variational autoencoders are used to deconstruct and reconstruct the multivariate time series data to learn the normal patterns. The anomaly inversion mechanism identifies concept drift in the anomaly prediction stage by dynamically updating the classification thresholds. Intelligent anomaly analysis then provides anomaly dimensions to help engineers better analyse the anomalous behaviour. Our experiments show that OOA-UADS performs satisfactorily. On seven streaming datasets, OOA-UADS outperforms 11 baselines in terms of AUC and provides state-of-the-art F1 scores on three batch datasets.

Zeroing neural network (ZNN) with different activation functions (AFs) for finding zero-result of time-varying quadratic programming (TVQP) with no noises are revisited. To improve the convergent speed of the ZNN and resist various noises occurred in the real application, two robust terminal recurrent neural network (RTRNN) models by adding two different AFs are presented for the exact solution of the TVQP problem facing various noises. The appearing advantage of the prespecified time of the RTRNN model is independent of the initial status of a generated system and the convergent time can be accelerated in advance, which is much superior than the finite-time performance with regard to the initial status. In addition, the prespecified convergent time of the RTRNN is mathematically discussed in detail under external noises. Simulated comparisons between the proposed RTRNN and the state-of-the-art neural networks substantiate the predefined time performance and strong robustness.

Aiming at low efficiency of agricultural machinery scheduling, this paper proposes an adaptive immune-following algorithm (AIFA) based on immune algorithm and artificial fish swarm algorithm. The adaptive crossover operator is used to accelerate convergence, and adaptive mutation operator ensures good diversity of population. After the adaptive evolution operations are performed, the following operator based on the following behavior of artificial fish swarm algorithm is embedded into the algorithm, which improves the convergence precision and obtains the promising optimization results. Experiments on scheduling considering the breakdown of agricultural machinery are performed based on multiple regions and multiple agricultural machineries. Compared with the immune algorithm and genetic algorithm, the simulation results demonstrate that AIFA can converge faster and achieve a better optimal solution.

Large and complex machines are the backbone of manufacturing and will remain key to industrialization for the foreseeable future as will leveraging essential technologies, such as the Internet of Things. However, efforts to make this machinery as smart as it could be are falling behind the curve. Data quality is often too low or too heterogeneous for useful analytics making maintenance troublesome. Additionally, many machines are still dumb with no uniform way to extract data or monitor their operation. In part, the success of future concepts like intelligent manufacturing, cyber physical systems, and industry 4.0 depends on solving these problems. Hence, this article presents FADA the groundwork for an ontology and 3-layer cloud-fog-edge architecture for large and complex machines that places data acquisition and IoT at the forefront. The ontology provides a flexible framework for standardizing data. The fog nodes acquire data directly from smart machines, while the edge nodes harvest data from dumb equipment through a recognition model. The fog nodes are flexible and multi-threaded to provide faster higher-performance computing power. To evaluate the proposed architecture and concepts, we implemented FADA in two factory-based testbeds: one with IoT-enabled equipment, the other with mostly dumb machines. The response times and influence rates recorded are promising and indicate that the system is highly adaptable to many different scenarios. We also conducted comparative experiments between FADA and a conventional data acquisition system to compare the occupied disk space, processing time, and data uploading time, which show that the FADA can save 2.9 TB of disk space per day, and reduce the server’s processing time by 184.8 ms per time over the conventional data acquisition system(CDAS), when 20000 fog nodes simultaneously access the server. The results show improvements by FADA in all metrics.

Cities are becoming smarter because of recent advances in artificial intelligence and the Internet of Things. However, heterogeneous data source in smart cities are continuously producing low-quality data, and ever-growing applications have greater real-time requirements. Therefore, this article proposes a data stream cleaning system (named DSCS) using edge intelligence to utilize the advantages of cloud servers and edge devices. The DSCS in edge nodes consists of a dynamic protocol interpreter, a structure parser, and a cleaning model activator. Meanwhile, a cloud server, which has pools of protocol and structured programs and cleaning models, supports the edge nodes to adapt massive heterogeneous data sources. To validate the proposed data cleaning system, we applied it to two scenarios: monitoring the injection molding machines, and base stations. The DSCS can have a stable processing time when the number of accessed edge devices is increased, as well as a good cleaning effect.

Time synchronization is an important function for networked control systems. At present, most of the time synchronization methods are based on proportional–integral controllers. These methods rely on the time offset calculated by the sampling data and are, therefore, vulnerable to sampling noise. This study introduces a stochastic search method and proposes a novel time synchronization method with antinoise ability, especially pulse noise. In this article, mathematical models of the proposed method are given, and its behavior is analyzed through simulation. The performance of the proposed method is verified by a comparative experiment, and the experiment results show that the proposed method has a faster convergence speed and better accuracy in a switched networked control system.

To obtain the solution for time-varying quadratic programming (QP), a time controlling neural network (TCNN) is presented and discussed. The traditional recurrent neural networks provide a prospect for real-time calculations and repeatable trajectory control of the mobile manipulators due to its high executing processing and nonlinear disposal ability. However, the convergent time is still a considerable point for the solution of a dynamic system dealing with synchronism and robustness. In this note, a TCNN model by incorporating an initial rectified term is applied to solve the online calculation problems and the convergent time can be controlled in advance. Theoretical analyses on stability, prespecified time and convergence are rigorously clarified. Finally, effectiveness and precision of the TCNN model for the solution of a QP example have been verified. In addition, a repetitive trajectory planning for a three-wheel manipulator is introduced to demonstrate the superiority of the TCNN.

Environments built from the edge-based Industrial Internet of Things (IIoT) are maelstroms of information that continuously flows between heterogeneous data objects, such as sensors and devices, and edge nodes. However, the explosive growth in the number of IIoT data objects connected to edge nodes generally results in significant computation and storage requirements, which exceed those of resource-constrained edge nodes. The problem mentioned above causes a major concern related to efficient memory usage and faster communication processing in data acquisition. Therefore, in this article, we propose a multibuffers multiobjects (MBMOs) architecture to support the parallel delivery of multiple data objects to multiple variable-length buffer blocks. Furthermore, a mathematical model is established, aiming to find the optimum buffer according to the size of each communication data packet. For the aforementioned matching problem, a spatiotemporal resource allocation algorithm is designed to maximize memory usage while minimizing communication processing time. We implement our MBMO in a monitoring system in which embedded programmable logic controllers (ePLCs) serve as the edge nodes. The analyses illustrate that memory usage and time efficiency are improved greatly with the utilization of MBMO.

Regarding programmable logic controller (PLC) development, improving programming efficiency and encouraging code exchange and software reuse are necessary to increase the productivity and safety of smart manufacturing and industrial automation. However, differences between implementations are significant even though all manufacturers claim to conform to the IEC 61131-3 and IEC 61131-10 standards, resulting in incompatibilities inside heterogeneous systems, preventing projects from interoperating between vendors. In this article, we present an approach that utilizes an ontology-based semantic model named OntoPLC to enable automatic porting of PLC projects between development environments and also prevent significant information loss during the translation process. High-level semantics of PLC projects are added into OntoPLC including software resources, which can be required for software reuse leveraging semantic query. To demonstrate the usefulness of the ontology model, the proposed methodology is applied to a turbine-driven boiler feed pump control module.

The merging boundaries between edge computing and deep learning are forging a new blueprint for the Internet of Things (IoT). However, the low-quality of data in many IoT platforms, especially those composed of heterogeneous devices, is hindering the development of high-quality applications for those platforms. The solution presented in this article is intelligent data collaboration, i.e., the concept of deep learning providing IoT with the ability to adaptively collaborate to accomplish a task. Here, we outline the concept of intelligent data collaboration in detail and present a mathematical model in general form. To demonstrate one possible case where intelligent data collaboration would be useful, we prepared an implementation called adaptive data cleaning (ADC), designed to filter noisy data out of temperature readings in an IoT base station network. ADC primarily consists of a denoising autoencoder LSTM for predictions and a four-level data processing mechanism to perform the filtering. Comparisons between ADC and a maximum slop method show ADC with the lowest false error and the best filtering rates.

Facing globalized competition, there have been increasing requirements for safety and efficiency in smart factories, where the industrial Internet of Things can enable the monitoring of equipment's status and the detecting of faults before they go critical. Regarding cloud computing, data-driven methods running at clouds are adopted to train the model with a large amount of raw data at the beginning, then end machines upload their real-time readings to the cloud center for processing. However, this incurs considerable computational costs and may sometimes bear a severe delay. In this article, we consider a hierarchical structure where edge-PLCs are employed to gather sensed data locally and reduce communication costs. Since a single fault may be related to multiple influencing features, we want to first minimize the number of features that need to determine a fault, then try to find out the minimal set of edge-PLCs which can cover all key features so as to save the deployment cost. We propose a random-forest-based method to handle the features selection problem, and then the selection of edge-PLCs by solving the set coverage problem. Through the simulation on real data trace, we compare our method with other artificial-intelligence-based methods, such as the logistics regression model and its extensions. The results prove the efficiency and performance of the proposed method, which reaches or even exceeds the accuracy of methods using the full set of data.

Edge computing has enabled extensive reasoning capabilities at the edge of the network. Edge nodes have direct access to edge sources (e.g., sensors, devices) to read and process relevant data, while each sensor/device manufacturer uses their bespoke protocols in the Internet of Things (IoT) environment, hence, addressing compatibility problems to ensure seamless access has become a cumbersome task. A conventional solution is to save all sensor/device protocols in an edge node, however, which brings higher redundancy and hardware expenditure. Hence, we propose a dynamic edge access system (DEA) built on a collaborative IoT architecture. It solves the compatibility issues with low redundancy by customizing access programs. Besides, embedded programmable logic controllers are designed as edge nodes and have been applied in around 78 000 base stations. After implementing DEA, the analysis illustrates that the average improvement of every station's protocol utilization is 96.3% and 66.6%, while the memory footprint is decreased to 4.5% and 33.3% compared with the conventional solution and global dynamic reconfiguration method.

General programmable logic controllers (PLCs) are difficult to adapt to changeable applications, especially for those with special motion control; this has seriously affected the development efficiency. This paper presents the concept of the embedded PLC (ePLC), whose hardware structure can be customized according to actual requirements. We proposed a three-layer architecture, and its customizable application layer could be compiled in real time. Correspondingly, the PLC program was divided into an engine program, control program, and customizing program. The description and compilation method of the customizing program was provided. We presented a customized winding machine language based on the proposed ePLC software structure. This was implemented in an automatic winding machine, which was easier to use compared with some widely used languages (e.g., G-Code).

The acquisition and analysis of datasets including multi-level omics and physiology from non-model species, sampled from field populations, is a formidable challenge, which so far has prevented the application of systems biology approaches. If successful, these could contribute enormously to improving our understanding of how populations of living organisms adapt to environmental stressors relating to, for example, pollution and climate. Here we describe the first application of a network inference approach integrating transcriptional, metabolic and phenotypic information representative of wild populations of the European flounder fish, sampled at seven estuarine locations in northern Europe with different degrees and profiles of chemical contaminants. We identified network modules, whose activity was predictive of environmental exposure and represented a link between molecular and morphometric indices. These sub-networks represented both known and candidate novel adverse outcome pathways representative of several aspects of human liver pathophysiology such as liver hyperplasia, fibrosis, and hepatocellular carcinoma. At the molecular level these pathways were linked to TNF alpha, TGF beta, PDGF, AGT and VEGF signalling. More generally, this pioneering study has important implications as it can be applied to model molecular mechanisms of compensatory adaptation to a wide range of scenarios in wild populations.