ECoLAD: Deployment-Oriented Evaluation for Automotive Time-Series Anomaly Detection

2026-03-11

AI Summary

Existing time series anomaly detection methods are predominantly evaluated under idealized workstation conditions, overlooking critical deployment constraints in automotive environments—particularly the need for predictable latency and limited CPU parallelism—rendering current leaderboards poor indicators of real-world feasibility. This work proposes ECoLAD, a novel deployment-oriented evaluation protocol that systematically assesses detectors under throughput constraints by constructing a descending ladder of computational resources, incorporating integer scaling rules and explicit CPU thread limits. Evaluations on automotive telemetry data (anomaly rate ≈ 0.022) and public benchmarks reveal that lightweight traditional methods maintain high coverage and effective performance across the full throughput spectrum, whereas several deep learning models lose deployment viability before exhibiting significant accuracy degradation.

Technology Category

• Machine Learning: Evaluation and Analysis
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Evaluation and Analysis

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Security and Privacy: Large-scale security measurements

Abstract

Time-series anomaly detectors are commonly compared on workstation-class hardware under unconstrained execution. In-vehicle monitoring, however, requires predictable latency and stable behavior under limited CPU parallelism. Accuracy-only leaderboards can therefore misrepresent which methods remain feasible under deployment-relevant constraints. We present ECoLAD (Efficiency Compute Ladder for Anomaly Detection), a deployment-oriented evaluation protocol instantiated as an empirical study on proprietary automotive telemetry (anomaly rate ${\approx}$0.022) and complementary public benchmarks. ECoLAD applies a monotone compute-reduction ladder across heterogeneous detector families using mechanically determined, integer-only scaling rules and explicit CPU thread caps, while logging every applied configuration change. Throughput-constrained behavior is characterized by sweeping target scoring rates and reporting (i) coverage (the fraction of entities meeting the target) and (ii) the best AUC-PR achievable among measured ladder configurations satisfying the target. On constrained automotive telemetry, lightweight classical detectors sustain both coverage and detection lift above the random baseline across the full throughput sweep. Several deep methods lose feasibility before they lose accuracy.

Problem

• time-series anomaly detection
• deployment constraints
• automotive telemetry
• compute efficiency
• throughput-constrained evaluation

Innovation

• deployment-oriented evaluation
• compute-constrained anomaly detection
• throughput-aware benchmarking
• automotive time-series
• efficiency ladder