|
| References | Arguments | Challenges | Application areas |
|
| [19] | (i) A need of automatic fault detection for large and complex systems | (i) Large dimensionality of monitored variables | Automotive engine test benches. Heavy diesel engine (caterpillar) |
| (ii) Address the restrictive safety and environmental regulations | (ii) High sampling rates |
| (iii) Nonstationary patterns |
| (iv) False alarms |
|
| [26] | (i) Take maintenance action in advance of actual failures | Physical models of the covered structures in normal and anomalous states are unavailable or of limited fidelity | Missile defense system structural components |
| (ii) Minimize downtime and use resources efficiently |
| (iii) Decrease costs and impact readiness of schedule-based preventive maintenance |
|
| [23, 32] | The need of resource-constrained monitoring of time-critical data streams where central collection of data is an expensive proposition | Monitoring fleet of vehicles and associated data streams in a resource-constrained environment | Vehicle (ford taurus car) and driver characterization |
|
| [22] | (i) Achieve optimal performance of machining process | Real-time monitoring | Cutting tool machines |
| (ii) Need of online cutting tool condition monitoring |
| (iii) Cost saving |
|
| [24] | Vehicle health monitoring is an area of interest for NASA in terms of vital subsystems on the spacecraft | (i) Analyze large, complex, multivariate time-series in near-real time | Spacecraft |
| (ii) The dynamics of the system cannot be modeled |
|
| [43] | (i) Increase the efficiency of monitoring | (i) Limited expert knowledge | Steel industry (metal sheet forming processes in rolling mills) |
| (ii) Minimize system down time for repair and maintenance | (ii) Fault patterns not predefined |
| (iii) Fault patterns cannot be simulated |
|
| [44] | (i) Reduce unscheduled machine down time | (i) Curse of dimensionality | Metal industry and car engines |
| (ii) Decrease repair costs | (ii) Ideal time lag estimation |
| (iii) Increase production efficiency | (iii) Inclusion of output (error) feedback |
| (iv) Structure identification (linearity versus non-linearity) |
| (v) Parameter estimation |
|
| [22] | (i) Detect deviations and monitor machine health status | (i) Fast-arriving data from multiple sensors | General framework for machine monitoring |
| (ii) Save damage costs | (ii) Rapid online and real-time analysis |
|
| [11] | (i) Manufacture products of high quality | Monitor data stream in real time | Hydraulic systems |
| (ii) Reduce the consequences of equipment failures in terms of time and cost |
|
| [17] | (i) Detecting failures at an early stage or foreseeing them before they occur is crucial for machinery availability | (i) Real-time monitoring | Hydraulic systems |
| (ii) Data prediction can reduce the consumption of communication resources in distributed data stream processing | (ii) Failures may occur suddenly (in short time) |
|
| [42] | Processing data streams from controllers and sensors is critical for monitoring the functional product in use | Scale up data analysis for handling huge amounts of equipment | Milling |
|
| [34] | Increasing product and process availability | Ability to search data streams while dealing with concept drift | Hydraulic systems |
|
| [33] | Increase the availability of industrial companies’ products | Monitor data stream in real time | Hydraulic systems |
|
| [35] | To achieve predictive failure management for fault-tolerant data stream processing | Providing lightweight failure prediction in an online and streaming setting | Software |
|
| [45] | The need of highly sophisticated supervisory and control schemes to satisfy a certain degree of performance when unfavorable conditions are occurring in critical infrastructure systems (CIS) | (i) Analytical models are not applicable | Drinking water network |
| (ii) Real-time monitoring |
|
| [36, 38] | To deliver quality services for industrial equipment by continuously monitoring its behavior | (i) On-board condition monitoring | Volvo CE wheel loaders |
| (ii) Real-time sensor analysis |
| (iii) Distributed data sources |
|
| [37] | Provide a framework and taxonomy of anomaly symptoms for low latency online anomaly detection | (i) Real-time anomaly detection in embedded system | Autonomous vehicle/Advanced driver-assistance systems (ADAS) |
|
| [40] | Detecting outliers in multiple concurrent data streams | (i) Parallel processing for outlier detection in data streams | Detecting contextual outliers |
|
| [39] | Analyzing data streams in industrial processes and industrial cyber-physical systems | (i) Provide scalable capability to visualize the results from the analysis of data streams to support industrial needs | Industrial analytics applications |
|
| [41] | A method to handle nonstationary and dynamic data streams where the distributions are altered with the time | (i) Real-time applications with time and memory constraints | Applied on standard datasets from literature |
|
| [42] | Utilizing data-driven models for anomaly detection in the industrial area | (i) Large-scale data sets | Metro do porto subsystems |
|
