The research, led by Raj Kumar Batchu of Amrita Vishwa Vidyapeetham together with colleagues at the National Kaohsiung University of Science and Technology, tackles a blind spot that has persisted for two decades. Since the early signature-based detectors of the 2000s, intrusion detection research has migrated toward deep learning architectures that achieve spectacular accuracy on benchmark datasets. Yet the authors point out that nearly all comparative studies report a single number, usually accuracy or F1-score, while ignoring the questions that actually decide whether a system survives in production: Can it run on a constrained edge device? Does it catch every category of attack or just the common ones? And what will it truly cost to operate over five years?

To answer those questions, the team built a multi-dimensional evaluation framework resting on three pillars. The first, Pareto Frontier Optimization, maps each candidate detector in a three-dimensional space of detection performance, computational efficiency and deployment feasibility, then identifies non-dominated solutions where no objective can be improved without sacrificing another. The second, a newly introduced metric called Attack Surface Coverage, compresses true positive rate, false positive minimization, adaptability to novel attacks and inference speed into a single weighted score. The third, a Total Cost of Ownership model, breaks down five years of deployment across 1,000 IIoT devices into five components: deployment, operations, incident response, scalability and compliance. A synthesis engine then fuses all dimensions into one composite ranking, with weights that organizations can tune to their own priorities.

The framework was stress-tested against three state-of-the-art intrusion detection models, all evaluated on the Edge-IIoTset 2022 dataset, a realistic benchmark containing 61 network traffic features and 15 distinct attack types spanning reconnaissance, exploitation and lateral movement. Model A was a hybrid convolutional neural network combined with gated recurrent units and long short-term memory layers, trained through asynchronous federated learning so that edge nodes keep their data local while a global model is aggregated centrally. Model B paired an autoencoder, which compresses network flows into low-dimensional representations, with a decision tree classifier, an unusually lightweight combination validated on an NVIDIA Jetson Nano. Model C ran parallel CNN and bidirectional LSTM branches topped with a self-attention mechanism, trained with SMOTE oversampling to handle class imbalance across rare attack categories.

The headline results upend the intuition that the most accurate model wins. On raw detection, the federated CNN-GRU-LSTM model was flawless, achieving 100 percent accuracy, precision, recall and F1-score with a measured zero false positive rate. But in the Pareto analysis, the humble decision tree with autoencoder dominated, earning a composite score of 0.999 against 0.333 for the federated deep model and 0.344 for the attention architecture. The reason is brutal arithmetic: the deep models carry footprints of 80 to 100 megabytes and training cycles measured in thousands of seconds, while the decision tree pipeline infers in roughly 0.185 milliseconds per sample with a compact model that edge hardware can actually host.

The Attack Surface Coverage scoring told a different story. Here the LSTM-CNN-Attention model led with 98.16 percent, edging out the federated model at 97.00 and the decision tree at 95.48. The attention mechanism, which re-weights the most salient temporal and spatial features of traffic sequences, gave the architecture an edge in generalizing to attack behaviors it had never seen, a property the authors encode as an architecture-informed adaptability score. The framework is candid that this sub-metric reflects a defensible relative ordering across architecture families rather than a directly measured zero-day detection rate, a transparency the authors extend to every formula, weight and cost assumption in the paper.

The economic dimension produced perhaps the most striking finding of all. Over a five-year horizon across 1,000 devices processing 10,000 flows each per day, the CNN-GRU-LSTM model achieved the lowest total cost of ownership at 1.385 million dollars, compared with 5.5 million for LSTM-CNN-Attention and a staggering 28.16 million for the decision tree with autoencoder. The paradox dissolves when the cost model is examined closely. Because the federated model’s false positive rate is exactly zero, its incident response cost, which multiplies every false alert by analyst investigation time and labor rates, is structurally zero. The lightweight decision tree, despite cheap hardware, generates enough false alarms at industrial traffic volumes to accumulate enormous analyst workload. A sensitivity analysis varying traffic volume, investigation time and labor rates confirmed that the federated model’s cost advantage is tied to its zero false positives rather than to any particular assumption.

When the synthesis engine combined all dimensions with its default weights, 30 percent detection, 25 percent security coverage, 20 percent cost, 15 percent deployment and 10 percent efficiency, the decision tree with autoencoder claimed the highest composite score at 78.85 percent, ahead of the federated model at 74.25 and the attention model at 71.98. Yet the authors emphasize that no single model dominates every criterion, and their sensitivity analysis shows the ranking shifts predictably as organizational priorities change. Weighting detection and cost favors the federated model; weighting edge deployment and efficiency favors the decision tree; weighting adaptability favors the attention architecture. The framework is deliberately parametric, letting a security team adjust the weight vector to match its own threat landscape and budget.

The study is honest about its limits. The authors did not independently retrain the models, drawing instead on previously published results, so the comparative conclusions should be treated as indicative rather than definitive, and rankings may vary with different datasets, weights and cost parameters. Still, the implications reach well beyond this single benchmark. For an industry where regulatory frameworks such as IEC 62443 demand auditable justifications for security decisions, and where black-box deep models can inflate compliance costs through required third-party validation, a transparent, reproducible selection framework that prices interpretability, alert fatigue and scalability into the decision could prove transformative. The full calculation logic has been released in a public repository, and the authors plan to extend the approach to broader families of lightweight algorithms and additional datasets.

What emerges is a deceptively simple message with far-reaching consequences: the best intrusion detector for a factory floor is not the one with the highest accuracy on a leaderboard. It is the one whose accuracy, coverage, hardware demands and five-year bill align with the specific constraints of the deployment. By turning an intuition into equations, weights and a tunable decision matrix, this work gives industrial defenders something they have rarely had, a defensible, quantified way to choose the shield that fits their world.

Subject of Research: Context-aware selection of intrusion detection systems for edge-based Industrial Internet of Things networks using multi-dimensional evaluation

Article Title: Multi-dimensional framework for context-aware IDS selection in edge-IIoT

Article References: Batchu, R. K., Padarthi, S. K., Nandanapu, R., Pandey, N., Malineni, R., & Shieh, C.-S. (2026). Multi-dimensional framework for context-aware IDS selection in edge-IIoT. Discover Artificial Intelligence, 6(1), Article 1473. https://doi.org/10.1007/s44163-026-02433-3

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02433-3

Keywords: intrusion detection systems, Industrial Internet of Things, edge computing, deep learning, federated learning, autoencoder, attention mechanism, total cost of ownership, Pareto optimization, cybersecurity, Edge-IIoTset dataset, multi-criteria decision making

The research team, led by Sikandar Abdul Qadir and Amjad Ali of King Fahd University of Petroleum and Minerals, built a Monte Carlo framework that runs 10,000 simulated ownership histories for each vehicle pair. Rather than fixing a single annual mileage or energy price, the model draws daily driving distances from a lognormal distribution fitted to Saudi survey data, in which 81.5 percent of drivers report travelling less than 100 kilometres per day. Maintenance costs, insurance premiums, and the growth rates of electricity and gasoline prices are likewise sampled from probability distributions. Energy prices follow a two-stage compound growth path, reflecting the observation that Saudi tariff revisions tend to occur roughly every five years. All costs are discounted at a fixed 5.5 percent rate, chosen for the Kingdom’s stable monetary environment and the peg of the Saudi riyal to the US dollar, and summed into a net-present-value total cost of ownership over ten years.

The choice of vehicles matters for policy relevance. Governments can only subsidise cars that are actually on sale, so the study paired the BYD Atto 3, priced at 99,900 riyals, against the Hyundai Elantra at 90,882 riyals, and the MG4 at 113,900 riyals against the Toyota Camry at 105,340 riyals. The electric models are more efficient by a wide margin: the Atto 3 travels 6.25 kilometres per kilowatt-hour and the MG4 6.86, while the Camry manages 17.2 kilometres per litre and the Elantra 15.5. Maintenance is modelled at 697 to 1,161 riyals per year for the EVs, roughly half the 1,626 to 2,500 riyals assigned to the combustion cars. The baseline assumes public charging at 0.99 riyals per kilowatt-hour, since nationally representative data on home charging are not yet available, and regulated gasoline at 2.18 riyals per litre. Notably, the model excludes a separate battery-replacement cost, because battery-ageing studies suggest modern packs retain roughly 79 to 87 percent of capacity after a decade.

Depreciation is where the economics turn. Because longitudinal resale data for electric cars in Saudi Arabia do not yet exist, the researchers applied mileage-based depreciation rates from international benchmarks: at 10,000 kilometres per year, EVs lose 15 percent of value annually against 8 percent for combustion cars, and the gap widens with mileage, reaching 22.21 percent versus 13.70 percent at 40,000 kilometres. Under this baseline scenario, which the authors call asymmetric depreciation favouring combustion vehicles, the gasoline cars dominate decisively. The Atto 3 costs 9.2 to 11.5 percent more to own over ten years than the Elantra across all mileage bands, and the MG4 runs 9.28 to 11.7 percent more than the Camry. At 10,000 kilometres per year, the Elantra’s levelised cost of driving is 11 percent cheaper per kilometre than the Atto 3’s. In the Monte Carlo classification, the combustion advantage exceeds 5 percent in 100 percent of simulations at every mileage level.

The picture shifts when the depreciation assumption changes. In a second scenario with identical depreciation rates for both powertrains, the results collapse into approximate cost parity: the EVs hold a slim 0.5 to 0.7 percent mean advantage at 10,000 kilometres per year, but the combustion cars claw back the lead as mileage rises, finishing 1.0 to 5.1 percent cheaper at higher utilisation. In a third, hypothetical scenario in which EVs retain value better than combustion cars, reflecting a maturing secondary market and tightening climate policy, the electric models win clearly at low and moderate mileage. The Atto 3 undercuts the Elantra by 3.9 to 12.8 percent in total cost, and the MG4 beats the Camry by 5.0 to 13.8 percent, though the advantage narrows and converges toward parity at the highest mileage band. The probabilistic classification confirms the pattern: at 10,000 kilometres per year, the EV advantage exceeds 5 percent in essentially all simulations, but at 40,000 kilometres the outcome is overwhelmingly parity rather than a decisive win.

Because Saudi summers are brutal, the team also stress-tested energy consumption with penalties of 10, 20, and 30 percent, consistent with field evidence that real-world consumption in extreme heat can exceed laboratory figures by 15 to 40 percent. Continuous air-conditioning and battery cooling raise electricity use, and one cited study measured a 28.4 percent increase in battery-electric consumption at 46 to 50 degrees Celsius relative to 26 to 30 degrees. The symmetric penalty applied to both powertrains erodes but does not overturn the scenario results. Under the EV-favourable depreciation scenario, the Atto 3 still holds advantages of 12,711, 8,960, and 4,034 riyals at the three lowest mileage bands even with the full 30 percent penalty, while the MG4 retains advantages of 15,089, 11,145, and 6,008 riyals. Under combustion-favourable depreciation, the gasoline cars remain cheaper throughout, with the EV gap reaching roughly 22,000 riyals at the joint worst case.

The most actionable findings come from the policy sensitivity analysis, which isolates fuel prices, charging tariffs, and purchase rebates. If Saudi gasoline prices, currently regulated at 2.18 riyals per litre, were to converge toward global averages of 3.5 to 4.0 riyals, the economics flip for anyone driving 20,000 kilometres or more per year: cost parity arrives at 3.5 riyals per litre for the Atto 3 and 3.75 for the MG4 at 20,000 kilometres, and the EVs dominate outright at 30,000 and 40,000 kilometres. Charging tariffs are equally decisive. At or below 0.30 riyals per kilowatt-hour, roughly the residential rate used here as a theoretical lower bound, EVs win for all drivers above 10,000 kilometres annually; at 0.50 riyals they still win for medium and high utilisation; but at the 0.99-riyal public-charging baseline they lose across the board. Purchase rebates deliver a third lever: a 15,000-riyal rebate achieves approximate cost parity for annual driving up to 30,000 kilometres, and 20,000 riyals or more secures a clear EV advantage, particularly relevant given that the Saudi average for private cars is about 23,942 kilometres per year.

The authors are careful about what their scenarios do and do not prove. The three depreciation cases are structural sensitivity tests, not forecasts, and actual resale values will depend on battery health, warranty coverage, new-vehicle pricing, and the maturity of a Saudi used-EV market that barely exists yet. Insurance costs are modelled with an identical range for both powertrains, a simplification that should be revised as Saudi insurers accumulate electric-vehicle claims data. The selected models also differ in body style and size, so the findings compare specific market alternatives rather than universal properties of the two drivetrains. The study likewise stops at the consumer level: fiscal affordability, grid impacts, and emissions outcomes, including earlier evidence that each additional 1 percent EV share in the Saudi power system cuts total emissions by about 0.5 percent on average, sit outside the model’s scope.

Even with those caveats, the policy sequencing that emerges is unusually concrete. The authors recommend starting with purchase rebates to build initial market volume, then deploying measures that stabilise residual values, such as certified pre-owned EV programmes, extended battery warranties, and subsidies for second-hand electric cars, before gradually reforming fuel prices and lowering public-charging tariffs. They also warn, citing China’s experience, that prolonged rebate dependence can distort the vehicle market, so incentives should be recalibrated as the market matures. For now, the combustion car remains the rational purchase for most Saudi drivers, but the analysis shows the gap is not a law of nature. It is a function of resale confidence, electricity pricing, and fuel reform, and each of those levers is within reach of policymakers pursuing Vision 2030 and the Kingdom’s 2060 carbon-neutrality pledge.

Subject of Research: Total cost of ownership comparison of electric and internal combustion vehicles in Saudi Arabia

Article Title: Cost analysis of electric and internal combustion vehicles in Saudi Arabia: the role of depreciation, energy pricing, and purchase rebates

Article References: Abdul Qadir, S., Ali, A., Islam, M. T., Shahid, M., & Atawi, I. E. (2026). Cost analysis of electric and internal combustion vehicles in Saudi Arabia: the role of depreciation, energy pricing, and purchase rebates. Results in Engineering, 32, Article 113203. https://doi.org/10.1016/j.rineng.2026.113203

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113203

Keywords: electric vehicles, total cost of ownership, Saudi Arabia, depreciation, Monte Carlo simulation, energy pricing, purchase rebates, levelised cost of driving, charging tariffs, fuel prices, Vision 2030, transport decarbonisation