The increasing prevalence of Unmanned Aerial Vehicles (UAVs) in modern warfare, dramatically highlighted by recent conflicts, has made anti-drone operations a critical domain of military competition. Defending against drone swarms, particularly low, slow, and small (LSS) targets, presents significant challenges in terms of cost-effectiveness, operational risk, and command complexity. Consequently, the design and optimization of combat schemes are paramount. A anti-drone scheme serves as the foundational blueprint for translating command intent into coordinated action, directly impacting the efficiency of the defensive system and the ultimate outcome of an engagement. However, evaluating these schemes is inherently difficult due to the multitude of involved assets (kinetic, non-kinetic, sensors, command systems) and the complex interplay of factors ranging from effectiveness and cost to risk and decision conditions.
This paper addresses this challenge by proposing a comprehensive evaluation framework based on the Analytic Hierarchy Process (AHP). The core of this research lies in constructing a hierarchical index system that decomposes the overarching goal of an optimal anti-drone scheme into measurable criteria and indicators. By leveraging expert judgment to weigh these factors and combining the results with data from military simulation, we enable a dimensionless comparative assessment of different anti-drone schemes. This methodology not only solves the problem of deriving an overall scheme evaluation from disparate, measurable indicators but also illuminates which factors most significantly influence operational effectiveness, thereby guiding more targeted and feasible optimization efforts.
1. Constructing the Hierarchical Evaluation Index System for Anti-Drone Schemes
A robust evaluation of anti-drone combat schemes necessitates a structured system that can encompass the multifaceted nature of such operations. We adopt a three-layer hierarchy: the Objective Layer, the Criterion Layer, and the Indicator Layer. This structure facilitates a logical progression from the ultimate goal to specific, quantifiable metrics.
1.1 Objective Layer: Defining the Strategic Aim
The Objective Layer represents the ultimate purpose of the anti-drone scheme. It provides the strategic direction and defines what constitutes mission success. For anti-drone operations, objectives can vary, such as: the complete destruction of all incoming drones, the attrition of a specific percentage of the swarm, the repulsion of the drone attack, or the neutralization of the threat via deception and electronic warfare leading to forced landing or diversion. The evaluation of any scheme is fundamentally tied to how well it achieves this pre-defined top-level objective.
1.2 Criterion Layer: The Paradigm for Scheme Assessment
The Criterion Layer breaks down the overarching objective into universal assessment dimensions. These criteria form the evaluative paradigm, allowing for a standardized comparison across diverse anti-drone schemes. Based on analysis of operational requirements and expert consultation, we identify four primary criteria:
- Combat Task Completion (CTC): Measures the degree to which the core tactical goals (linked to the Objective Layer) are fulfilled.
- Severity of Operational Cost (SOC): Evaluates the expenses and losses incurred by the friendly (Blue) forces in executing the scheme.
- Risk of Operational Action (ROA): Assesses the vulnerability and potential for failure during the operation.
- Decisive Conditions for Victory (DCV): Focuses on factors that critically influence the final battle outcome, often related to force preservation and critical asset survival.
1.3 Indicator Layer: Quantifiable Measures
The Indicator Layer provides the tangible, measurable variables that operationalize the criteria. These indicators are directly tied to simulation outputs or operational data. Their design depends on the granularity of the analysis. A detailed simulation allows for finer, more numerous indicators, enabling more precise optimization. The following table outlines a possible set of indicators for a anti-drone scheme evaluation, linked to their parent criteria.
| Criterion | Indicator Code | Indicator Name | Description & Measurement |
|---|---|---|---|
| Combat Task Completion (CTC) | C11 | Number of Enemy UAVs Destroyed/Neutralized | Total count of Red UAVs killed or mission-aborted. |
| C12 | Enemy UAV Attrition Rate | Ratio of neutralized Red UAVs to the total number engaged. $$ C12 = \frac{\text{Number of Neutralized Red UAVs}}{\text{Total Red UAVs in Attack}} $$ | |
| Severity of Operational Cost (SOC) | C21 | Blue Entity Losses | Number of Blue defensive assets (interceptors, sensors) destroyed. |
| C22 | Material Consumption Rate | Percentage of ammunition, energy, or other consumables expended. | |
| C23 | Total Mission Duration | Time from threat detection to mission end. | |
| Risk of Operational Action (ROA) | C31 | Blue Entity Loss Ratio | Weighted average ratio of destroyed Blue assets per type, reflecting their importance. |
| C32 | Red Precision Strike Probability | Estimated probability of Red’s advanced weapons successfully hitting Blue targets. | |
| Decisive Conditions for Victory (DCV) | C41 | Number of Blue Critical Assets | Count of high-value units (e.g., command centers, key radars). More assets may increase survivability through dispersion. |
| C42 | Blue Protection Level for Critical Assets | A composite score (0-1) based on defense layers and weapon density around critical assets. | |
| C43 | Red Probability of Destroying Critical Assets | Estimated probability (0-1) that Red forces can neutralize Blue’s critical assets. |
2. Methodology: Applying AHP for Weight Determination
The Analytic Hierarchy Process (AHP) is chosen for its suitability in solving multi-criteria decision-making problems with a hierarchical structure. It systematically evaluates the relative importance of factors at each level, ultimately deriving the global weight of each bottom-level indicator relative to the overall objective. This is crucial for anti-drone scheme optimization, as it clarifies the impact of specific measures on the final outcome.
2.1 Constructing Judgement Matrices and Calculating Weights
The process begins with constructing pairwise comparison matrices at each level of the hierarchy. Experts in anti-drone warfare are asked to compare the importance of elements using the standard Saaty scale (1-9). For example, a judgement matrix for the Criterion Layer (B) relative to the Objective Layer (A) might be constructed from expert surveys.
The fundamental scale for judgements is defined such that a value of 1 indicates equal importance, while 9 indicates extreme importance of one element over another. The judgements for the Criterion Layer are synthesized into a matrix \( A \):
$$
A = (a_{ij})_{n \times n}, \quad \text{where } a_{ji} = \frac{1}{a_{ij}}
$$
For illustration, the aggregated judgement matrix for the four criteria might be as shown below. The question answered is: “For evaluating an optimal anti-drone scheme, how much more important is Criterion i compared to Criterion j?”
| A / Criteria | CTC | SOC | ROA | DCV |
|---|---|---|---|---|
| CTC | 1 | 5 | 3 | 1/3 |
| SOC | 1/5 | 1 | 1/2 | 1/5 |
| ROA | 1/3 | 2 | 1 | 1/2 |
| DCV | 3 | 5 | 2 | 1 |
The relative weight vector \( W \) for the criteria is obtained by solving the eigenvalue problem \( AW = \lambda_{max}W \), where \( \lambda_{max} \) is the principal eigenvalue. The weights are then normalized. A critical step is checking consistency using the Consistency Ratio (CR):
$$
CI = \frac{\lambda_{max} – n}{n – 1}, \quad CR = \frac{CI}{RI}
$$
where \( RI \) is the Random Index. A \( CR < 0.10 \) is acceptable. This process is repeated for all sub-criteria relative to their parent criterion. After consistency verification and normalization, the final global weights for all indicators are calculated by multiplying the weight of an indicator by the weights of all its parent criteria up to the objective layer.
Based on this process, the derived global weights for our anti-drone evaluation hierarchy are summarized below. Note that CTC and DCV are positive indicators (higher value is better), while SOC and ROA are negative indicators (lower value is better).
| Criterion (Weight) | Indicator | Local Weight | Global Weight |
|---|---|---|---|
| CTC (0.31) | C11: UAVs Destroyed | 0.17 | 0.0527 |
| C12: Attrition Rate | 0.83 | 0.2573 | |
| SOC (0.07) | C21: Entity Losses | 0.65 | 0.0455 |
| C22: Consumption Rate | 0.23 | 0.0161 | |
| C23: Mission Duration | 0.12 | 0.0084 | |
| ROA (0.16) | C31: Entity Loss Ratio | 0.75 | 0.1200 |
| C32: Red Strike Probability | 0.25 | 0.0400 | |
| DCV (0.46) | C41: # Critical Assets | 0.28 | 0.1288 |
| C42: Protection Level | 0.59 | 0.2714 | |
| C43: Red Destruction Prob. | 0.13 | 0.0598 |
3. Case Study: Comparative Evaluation of Anti-Drone Scenarios
3.1 Scenario Descriptions
To demonstrate the evaluation methodology, we consider a Blue force defending a high-value site against a Red UAV swarm attack. The Red threat is fixed. Three different Blue anti-drone scheme configurations (Scenarios) are simulated and compared.
- Scenario 1 (Baseline): A modern, layered defense. Sensors include long-range radar, low-altitude gap-filler radar, and IR/EO systems. Interceptors comprise traditional systems (SAMs, gun-missile combinations) and novel effectors (tactical lasers, high-power microwave weapons).
- Scenario 2 (Degraded Sensors): Same as Scenario 1, but the detection range of all sensors is reduced by 50%.
- Scenario 3 (No Novel Effectors): Same as Scenario 1, but the laser and microwave weapons are removed, relying solely on traditional kinetic interceptors.
The simulations were conducted on a professional military simulation platform (Xsim), modeling the engagement dynamics to generate data for the indicators in Table 1.
3.2 Simulation Results and Normalized Assessment
The simulation outputs for key indicators across the three scenarios are tabulated below. This raw data forms the basis for the comparative assessment.
| Indicator | Scenario 1 (Baseline) | Scenario 2 (Sensors -50%) | Scenario 3 (No Lasers/Microwave) |
|---|---|---|---|
| Combat Task Completion (CTC) | |||
| C11: UAVs Destroyed | 167 | 163 | 162 |
| C12: Attrition Rate | 1.000 | 0.976 | 0.970 |
| Severity of Operational Cost (SOC) | |||
| C21: Blue Entity Losses | 4 | 6 | 5 |
| C22: Consumption Rate | 0.35 | 0.42 | 0.33 |
| Risk of Operational Action (ROA) | |||
| C31: Blue Loss Ratio | 0.0678 | 0.1017 | 0.0962 |
| C32: Red Strike Probability | 0.0240 | 0.0359 | 0.0299 |
| Decisive Conditions for Victory (DCV) | |||
| C42: Protection Level | 0.95 | 0.73 | 0.82 |
*Note: Indicators C23, C41, and C43 were constant across scenarios and thus do not differentiate them.
The overall score \( S_k \) for each scenario \( k \) is calculated by aggregating the weighted normalized indicator values. For positive indicators, normalization is \( v’ = v / v_{baseline} \). For negative indicators (SOC, ROA), normalization is \( v’ = v_{baseline} / v \). This ensures a higher score is better. The score is computed as:
$$
S_k = \sum_{i=1}^{n} (w_i \times I’_{k,i})
$$
where \( w_i \) is the global weight of indicator \( i \), and \( I’_{k,i} \) is its normalized value for scenario \( k \). Using Scenario 1 as the baseline (score = sum of its relevant global weights = 0.540), the scores for Scenarios 2 and 3 are calculated and then all are normalized for easy comparison.
| Scenario | Calculated Score | Normalized Score (Baseline=1.00) |
|---|---|---|
| Scenario 1 (Baseline) | 0.540 | 1.000 |
| Scenario 2 (Sensors -50%) | 0.257 | 0.477 |
| Scenario 3 (No Novel Effectors) | 0.303 | 0.562 |
4. Analysis and Insights from the Anti-Drone Scheme Evaluation
The quantitative evaluation yields clear insights into the performance and trade-offs of different anti-drone architectures.
4.1 Overall Scheme Superiority: The baseline scenario (Scenario 1), integrating modern sensors with a mixed traditional/novel interceptor fleet, is decisively superior, with a normalized score 75% higher than Scenario 2 and 44% higher than Scenario 3. This underscores the value of a holistic, modernized anti-drone system. The primary drivers of this superiority are found in the Decisive Conditions for Victory (DCV) and Risk of Operational Action (ROA) criteria. The advanced system significantly enhances the protection of critical assets (C42) and reduces the force’s vulnerability to precise strikes (C31, C32), factors which carry high weight in the overall assessment. The reduction in operational cost (SOC) is also notable, though less pronounced in this specific engagement scale.
4.2 Relative Impact of Subsystem Degradation: Comparing the two degraded scenarios reveals critical vulnerabilities. Degrading sensor performance (Scenario 2) has a more catastrophic effect on overall score than removing novel terminal effectors (Scenario 3). This is powerfully evident in the sharp increase in operational risk (ROA) and cost (SOC) when sensors are impaired. A blinded anti-drone system cannot effectively employ even its most advanced weapons. However, the analysis also shows that novel effectors (lasers, microwaves) contribute disproportionately to the “Decisive Conditions for Victory” by providing a highly effective, last-layer defense for critical assets (raising C42). They act as an indispensable “close-in shield,” complementing the longer-range “spear” of traditional interceptors.
4.3 The Interconnected Nature of Modern Anti-Drone Warfare: While the “Combat Task Completion” (attrition rate) showed less variation in this limited-scale simulation, this is a conditional result. In reality, the increased costs and risks in Scenarios 2 and 3 directly degrade the sustainability and survival of the defensive network, which would inevitably compromise task completion in a prolonged or larger-scale engagement. This highlights a core tenet of the proposed evaluation method: modern anti-drone combat is a complex system-of-systems endeavor. Factors like cost and risk are not independent side-notes but are intrinsically linked to and predictive of long-term effectiveness. An evaluation focusing solely on kill ratios would miss this crucial dynamic, potentially leading to fragile or unsustainable scheme designs.
5. Conclusion
This research establishes a systematic framework for evaluating and optimizing anti-drone combat schemes. By combining the structured decision-making of the Analytic Hierarchy Process with the empirical data from military simulation, the method provides a comprehensive, dimensionless metric for comparing disparate schemes. It successfully translates a multitude of measurable, low-level tactical indicators into a coherent assessment of overall scheme effectiveness, incorporating crucial dimensions of cost, risk, and decisive conditions alongside raw performance.
The case study demonstrates the practical utility of this approach. It not only ranks different anti-drone configurations but also diagnoses the specific criteria and indicators responsible for performance gaps. For force planners, this illuminates priority areas for investment—in this case, underscoring the fundamental importance of robust sensor networks while also validating the unique value of novel effectors for terminal defense. The methodology acknowledges the interconnectedness of modern anti-drone systems, where the degradation of one element (e.g., sensors) cascades into increased risk and cost, ultimately threatening mission success.
While the AHP-based method is powerful, its reliance on expert judgement for pairwise comparisons introduces a subjective element. Mitigating this requires careful selection of a diverse and knowledgeable expert panel. Future work could explore integrating this framework with dynamic optimization algorithms or machine learning techniques to automate the search for optimal scheme parameters under complex constraints, further advancing the science of anti-drone campaign planning.