Cost-Aware Conflict Triage for xApp Control Conflicts in Open RAN

This version of the paper has been accepted for presentation at IEEE WCSP 2026. It is subject to IEEE copyright and may be removed upon request.

Cost-Aware Conflict Triage for xApp Control Conflicts in Open RAN Abdul Wadud∗† , Fatemeh Golpayegani∗ , and Nima Afraz∗ ∗ School of Computer Science, University College Dublin, Dublin, Ireland † Bangladesh Institute of Governance and Management, Dhaka, Bangladesh

arXiv:2610.15310v1 [cs.NI] 11 Oct 2026

abdul.wadud@ucdconnect.ie, {fatemeh.golpayegani, nima.afraz}@ucd.ie

Abstract—Open Radio Access Network (Open RAN) enables multi-vendor deployments where concurrent RAN control applications (xApps) independently optimise mobility, load, coverage, and energy within the Near Real Time RAN Intelligent Controller (Near-RT RIC). When xApps act on overlapping control parameters, their isolated loops produce conflicting decisions that degrade the network Key Performance Indicators (KPIs). Existing conflict-evaluation methods estimate severity but do not answer whether the intervention is cost-justified. This paper proposes a cost-aware conflict triage framework that pairs ECDF-based severity evaluation with an event-level decision layer. The framework groups detected conflict records into episodes, extracts nine features, estimates per-mechanism costs and benefits, and assigns one of three actions: tolerate, defer, or mitigate. Evaluated in a simulator-driven Open RAN setting over 30 Monte Carlo runs with four concurrent xApps, the framework defers 66.8 %, mitigates 19.8 %, and tolerates 13.4 % of detected episodes, achieving a mean net utility of 0.132 per intervention, 3.9× higher than an always-mitigate baseline, by avoiding lowvalue interventions through cost-aware selectivity. Index Terms—Open RAN, xApp, conflict detection and mitigation, conflict triage, selective mitigation, cost-aware control, RIC, network utility

I. I NTRODUCTION PEN Radio Access Network (RAN) makes the radio access network more flexible by allowing operators to deploy multiple control applications from different vendors, called RAN control applications (xApps), inside the Near Real Time RAN Intelligent Controller (Near-RT RIC). Different xApps can independently optimise distinct aspects of the network, such as mobility robustness, energy saving, load balancing, or Quality of Service (QoS). However, this flexibility also creates a new challenge. Two or more xApps may attempt to influence the same control parameter, the same radio resource, or the same Key Performance Indicator (KPI) simultaneously. When that happens, conflicts arise, which can lead to network-wide performance degradation. Prior work covers conflict detection and taxonomy [1], rule-based and policy-based mitigation [2], [3], optimizationbased and learning-based resolution [4], [5], and proactive avoidance via scheduling or digital twins [6], [7]. However, most treat mitigation as automatic once a conflict is flagged. Notably, PACIFISTA [3] provides robust severity evaluation

O

Funding for this research is provided by the European Union’s Horizon Europe research and innovation program through the Marie Sklodowska-Curie SE grant under the agreement RE-ROUTE No 101086343.

but does not assess cost-justification, and QACM [4] selects the best mechanism but assumes mitigation is always warranted. The proposed triage layer fills this gap by adding the cost-awareness step that neither addresses. Mitigation carries real operational cost: processing overhead, signalling load, added latency, and disruption to ongoing control loops [8], [9]. Conflicts that produce only transient or minor degradation may not justify this cost, making whether to mitigate a distinct and necessary question alongside how to mitigate. Motivated by these observations, we introduce conflict triage as a decision layer for xApp conflict management. Once a conflict is detected, the controller chooses to tolerate (take no corrective action), defer (postpone mitigation and re-observe), or mitigate (intervene immediately). The key idea is simple: mitigate only when the expected benefit exceeds the expected cost. Built on top of PACIFISTA-style Empirical Cumulative Distribution Function (ECDF) severity evaluation [3], the proposed framework extracts nine episode-level features from detected conflict records, estimates the expected loss of inaction, compares it against per-mechanism cost and benefit across six candidates, and selects the action with the highest expected utility. The main contributions of this paper are: • We introduce conflict triage as a cost-aware decision layer for xApp conflict handling in Open RAN. • We combine an upstream PACIFISTA-style severity prior with an event-level utility model that supports tolerate, defer, and mitigate decisions. • We define a structured mitigation cost model that includes computation, signalling, latency, stability, and opportunity cost. • We demonstrate via 30-run Monte Carlo evaluation that the triage layer escalates high-severity, high-degradation episodes while deferring borderline cases, confirming internal consistency. II. S YSTEM M ODEL AND P ROBLEM F ORMULATION We consider an Open RAN deployment comprising one central unit (CU), multiple distributed units (DUs), and radio units (RUs), all managed through the Near-RT RIC. The network is observed at discrete time steps of duration ∆t , during which each xApp may issue at most one control action. Multiple xApps operate simultaneously on shared radio-control parameters through the Near-RT RIC.

Let X = {MRO, MLB, CCO, ES} denote the set of xApps considered in this work: Mobility Robustness Optimization (MRO) optimises handover parameters to reduce radio link failures; Mobility Load Balancing (MLB) redistributes load across cells by adjusting handover offsets; Capacity and Coverage Optimization (CCO) tunes transmit power for coverage and capacity trade-offs; and Energy Saving (ES) deactivates underloaded cells to reduce energy consumption. Each xApp issues control actions according to its own objective, while the resulting network behaviour is observed through KPI traces. Because the xApps do not optimise the same objective, an action taken by one xApp may degrade the KPI targeted by another xApp. This creates an xApp conflict. The system observes two main inputs over time: xApp action logs and KPI traces. From these inputs, the upstream monitoring layer identifies conflict evidence and associates it with a logical conflict event. We model one conflict event as an episode, e = (te , xi , xj , pe , ke , ce ), where te is the event time, xi is the instructing xApp, xj is the affected xApp, pe is the changed control parameter, ke is the degraded KPI, and ce ∈ {direct, indirect, implicit} is the conflict type. For each episode, the controller builds a feature vector, ϕe = [se , δe , ρe , ωe , κe , λe , ue , rb (e), rh (e)], where se ∈ [0, 1] is the PACIFISTA-derived severity prior for the instructing– affected xApp pair; δe ∈ [0, 1] is the normalised KPI degradation magnitude, computed as the mean relative KPI drop across all affected KPIs during the conflict window; ρe ∈ [0, 1] is the persistence score, reflecting the fraction of the observation window over which the degradation is sustained; ωe ∈ [0, 1] is the impact scope, capturing the proportion of affected cells or slices; κe ∈ [0, 1] is the operator-assigned KPI criticality weight; λe ∈ [0, 1] is the normalised system control-load state at the time of the episode; ue ∈ [0, 1] is the uncertainty estimate derived from the variance of the severity prior and KPI measurement noise; and rb (e) and rh (e) are the benign and harmful recurrence counts, respectively. The first seven features are normalised to [0, 1]; recurrence is represented by counts and bounded through the recurrence factor below. This design allows the controller to distinguish between conflicts that repeatedly self-resolve and conflicts that repeatedly cause damage. The controller can respond to an episode in three ways: tolerate it, defer the decision, or mitigate it. If mitigation is selected, one mechanism is chosen from the set of mitigation methods M. In this paper, M = {cancel, dampen, priority, nswf, eg, qacm}. The first three are unilateral action-modification mechanisms, while the last three are game-theoretic compromise mechanisms. cancel suppresses the conflicting action entirely when the safest choice is to block it. dampen reduces the strength of the action so that its negative impact is softened rather than fully removed. priority replaces the proposed action with a preferred alternative according to the controller’s resolution logic [2]. nswf applies a Nash social welfare objective to favour balanced outcomes across competing

xApps [5]. eg uses an egalitarian bargaining rule to promote fairness between conflicting objectives [5]. qacm provides a compromise-oriented mechanism that explicitly trades off benefit and cost under conflict [4]. These six mechanisms were chosen to cover both simple, low-overhead action modifications and cooperative, utility-driven compromises. Table II and Fig. 4 use the implementation labels cancel_latest, dampen_latest, and priority_override (shortened to override in Table II) for cancel, dampen, and priority, respectively. The opportunity cost component Copp (e, m) in Eq. 2 captures the foregone benefit of not applying an alternative mechanism, i.e., the utility difference between m and the next-best candidate. If a mechanism has no addressable action for a given ce (e.g., cancel applied to a conflict where no single action can be suppressed), its ηm,ce is set to zero, removing its loss-reduction benefit; its score still includes uncertainty and cost penalties. To decide whether intervention is worthwhile, we model the expected unmanaged loss, mitigation cost, residual loss after mitigation, mitigation benefit, and net utility as follows. The coefficients in these models can either be set by the Mobile Network Operator (MNO) according to operational priorities or calibrated offline from data. In this paper, we calibrate coefficient values using an offline multi-objective random-search procedure with Pareto filtering and weighted compromise selection. Lconf (e) = se δe (1 + βρ ρe + βω ωe (1) +βκ κe + βλ λe ) (1 + βh rh (e) − βb rb (e)) Cmit (e, m) = Ccomp (e, m) + Csig (e, m) + Clat (e, m) + Cstab (e, m) + Copp (e, m)
Lres (e, m) = 1 − ηm,ce Lconf (e) + ξm ue

(3)

B(e, m) = Lconf (e) − Lres (e, m)

(4)

U (e, m) = B(e, m) − Cmit (e, m)

(5)

(2)

Here, Lconf (e) represents the expected unmanaged loss, the cumulative KPI degradation that would accrue if the conflict is left unmitigated. Cmit (e, m) is the operational cost of applying mechanism m, comprising computation, signalling, latency, stability, and opportunity cost components (Eq. 2), reflecting the cost dimensions most consistently cited for real-time RIC control actions [8], [9] rather than an exhaustive taxonomy. Lres (e, m) is the remaining unmanaged loss after mitigation, B(e, m) is the mitigation benefit, and U (e, m) is the final net utility. The coefficients βρ , βω , βκ , and βλ control the relative influence of persistence, scope, KPI criticality, and load state, respectively, while βh and βb control the effect of harmful and benign recurrence. All β coefficients are non-negative (see Table II in Section IV for the calibrated values), and the recurrence factor (1 + βh rh (e) − βb rb (e)) is clipped to [0, 2] to prevent negative loss estimates under high

xApp Actions & KPI Traces

STAGE 1 Conflict Evaluation

PACIFISTA ECDF

severity prior

detected xApp conflict Conflict Records

Episode Aggregation

Feature Extraction

STAGE 3 Mitigation Execution

Unmanaged Loss

Per-Mechanism Scoring

Best Mechanism

re-observe

STAGE 2 Cost-Aware Triage (Proposed)

Triage Decision

↺ Defer

wise severity prior se ; detection supplies raw conflict records. Neither output alone triggers intervention: high severity with small observed degradation can still yield low utility in Eqs. 1– 5.

✓ Tolerate

⚙ Mitigate

Fig. 1. Proposed cost-aware conflict triage pipeline.

benign recurrence. The effectiveness parameter ηm,ce ∈ [0, 1] captures the expected fraction of unmanaged loss eliminated by mechanism m for conflict type ce ; it is calibrated offline for each mechanism–conflict-type pair using historical simulation data. The parameter ξm ≥ 0 is a mechanism-specific uncertainty penalty reflecting the reliability of each mechanism’s effect estimate (calibrated values in Table II). Thus, larger ηm,ce reduces the residual loss, while larger ξm makes the controller more cautious under uncertainty. The loss model is an operator-defined surrogate, not a physical law. The product se δe requires both prior conflict evidence and observed degradation; non-negative linear weights make contextual effects monotone and interpretable. The clipped recurrence factor bounds history effects. This structure supports transparent calibration but does not establish that the chosen form predicts actual network loss better than alternative models. The problem is therefore to assign a triage action to each conflict episode such that costly intervention is avoided when it is unnecessary, while harmful conflicts are still handled in time. Formally, the controller first selects the best mitigation candidate as m∗ (e) = arg maxm∈M U (e, m), and then maps the resulting utility to one of the three triage actions α(e) ∈ {tolerate, defer, mitigate}. The operational mapping from utility and uncertainty to the final triage action is defined in the proposed framework in Section III. III. P ROPOSED C OST-AWARE C ONFLICT T RIAGE F RAMEWORK Fig. 1 shows the proposed pipeline. The workflow has three stages: conflict evaluation, cost-aware triage, and mitigation execution. The first stage is upstream and provides conflict evidence. The second stage is the main contribution: rather than mitigating every detected conflict, the controller compares expected benefit with expected cost and chooses among tolerate, defer, and mitigate. The third stage applies the final decision through the conflict management component. A. Stage 1: Upstream Conflict Evaluation Action logs and KPI traces feed the upstream detector and PACIFISTA-style ECDF profiling. Profiling supplies a pair-

B. Stage 2: Cost-Aware Triage Records sharing an instructing–affected xApp pair within the aggregation window ∆t are grouped into an episode. The controller extracts ϕe , computes unmanaged loss once using Eq. 1, and scores each candidate using Eqs. 2–5. It retains the best mechanism m∗ (e) and its utility U ∗ (e) = U (e, m∗ (e)) for the decision stage. Benign and harmful recurrence enter the loss estimate separately, distinguishing self-resolving events from repeated damage. C. Stage 3: Triage Decision and Mitigation Execution Let ue denote the uncertainty of episode e. The thresholds τl and τh are the lower and upper utility thresholds, respectively, and uh is the uncertainty threshold. Thus, τl separates clearly non-beneficial cases from borderline ones, τh separates borderline cases from clearly beneficial ones, and uh limits mitigation under high uncertainty. The final triage action is then defined as   mitigate with m∗ (e), U ∗ (e) > τh ∧ ue ≤ uh ,    defer, τl ≤ U ∗ (e) ≤ τh , α(e) =  defer, U ∗ (e) > τh ∧ ue > uh ,    tolerate, U ∗ (e) < τl . (6)

Deferral is useful, but it should not continue forever. Therefore, the framework uses bounded deferral. If an episode is deferred repeatedly and its deferral count reaches the maximum allowed limit Dmax , the controller escalates the event and applies the best available mitigation. This keeps the system from ignoring persistent, harmful conflicts simply because early evidence was uncertain. The thresholds τl , τh , uh , and Dmax are calibrated alongside the model coefficients β using the offline random-search procedure described in Section II. Note that τl is set to a small negative value (−0.02), so that only episodes with clearly negative expected utility are assigned tolerate; episodes with slightly negative or borderline utility in [τl , 0] are deferred for re-observation rather than immediately dismissed. D. Computational Cost With episode features and calibrated coefficients available, loss estimation takes constant time. Scoring and selecting among M = |M| mechanisms takes O(M ) time and can use O(1) working storage by retaining the best score. Here M = 6; scoring uses arithmetic and comparisons without online training. This bound excludes upstream detection, trace aggregation, and execution of the selected mechanism, including any optimisation solver. End-to-end inference latency is not measured; the simulation’s observation interval does not establish a deployment latency guarantee.

Triage action distribution (n=30 runs, 95% CI)

TABLE I

80

T RIAGE THRESHOLD AND DEFERRAL PARAMETERS . 66.8%

70

Role Tolerate if U ∗ (e) < τl Mitigate if U ∗ (e) > τh , ue ≤ uh Defer if ue > uh Escalate after 2 deferrals (20 ms)

60

Action rate (%)

Parameter Value τl (lower threshold) −0.02 τh (upper threshold) 0.18 uh (uncertainty gate) 0.70 Dmax (max deferrals) 2

50 40 30

TABLE II L OSS - WEIGHT (β) AND UNCERTAINTY- PENALTY (ξm ) COEFFICIENTS . Coeff. βρ (persistence) βω (scope) βκ (KPI crit.) βλ (load) βh (harmful recur.) βb (benign recur.)

Value 0.35 0.25 0.40 0.25 0.18 0.12

Coeff. ξcancel ξdampen ξoverride ξnswf ξeg ξqacm

Value 0.10 0.06 0.08 0.08 0.07 0.05

IV. E XPERIMENTAL M ETHODOLOGY This section evaluates the proposed cost-aware conflict triage framework in a simulator-driven Open RAN setting. We use MATLAB 5G Toolbox to represent an O-RAN 7.2x architecture [10]. The evaluation assesses the internal behaviour of the policy layer; it does not constitute validation of a deployed Near-RT RIC or live E2 control loop. A. Simulation Setting For the experimental setup, we consider a four-xApp scenario presented in [4] that incurs direct and indirect conflict due to their xApp-parameter-KPI relationship dynamics. These four interacting xApps are: MRO, MLB, CCO, and ES. These xApps operate on shared radio-control parameters and generate action logs together with KPI traces, which form the inputs to the upstream conflict-evaluation stage and the proposed triage layer. For the evaluation, we use a fixed baseline network scenario with one CU, three DUs, and nine RUs under a 7.2x split. The radio profile is FR1 at 3.5 GHz with 100 MHz bandwidth and simplified fixed beamforming. Each run serves 100 users with mixed mobility patterns. The controller implementations are: DDQN-based MRO [11], supermodular load balancing for MLB [12], SVD-based power recommendation for CCO [13], and VBP-style energy saving for ES [12]. The DDQN-MRO agent completes 12,461 gradient updates (εfinal = 0.050), after which the trained policy is used in the evaluation. Table I lists the triage threshold and deferral parameters used in all experiments. These values were selected by the offline calibration procedure (24 candidates, 3 MC runs each, single Pareto-front winner) and kept fixed across all 30 evaluation runs. Table II lists the calibrated loss-weight and uncertaintypenalty coefficients. Calibration and evaluation use the same fixed topology and operating profile. B. Monte Carlo Protocol To evaluate stability, we run the full pipeline over 30 Monte Carlo trials with different random seeds and report the mean, standard deviation, and 95 % confidence interval for each metric. Unless otherwise stated, the triage layer uses the calibrated coefficient setting stated in Section III.

19.8% 20 13.4% 10 0

Tolerate

Defer

Mitigate

Fig. 2. Triage action distribution. Defer dominates at 66.8 %, confirming selective operation: the majority of detected conflicts are held for re-observation rather than immediately mitigated. V. R ESULTS AND D ISCUSSION

A. Overall Triage Behaviour and Stability The first result is that the proposed framework behaves as a selective decision layer rather than an always-on mitigation trigger. Across 30 Monte Carlo runs, the controller detects on average 118.3 conflict episodes per run (std 33.2, 95 % CI [105.9,130.7]). On average, defer is the dominant action at 66.8 % (CI [64.8, 68.7]), followed by mitigate at 19.8 % (CI [17.9, 21.7]) and tolerate at 13.4 % (CI [13.0, 13.8]), as shown in Fig. 2. Translating the rates to absolute counts: the framework defers a mean of 79.0 episodes, mitigates 23.4, and tolerates 15.8 per run. These 118.3 episodes arise from six structurally conflicting xApp pairs across all four xApps. MRO-facing pairs show the highest PACIFISTA severity priors (up to 0.241 for MRO– MLB) because MRO shares Tx Power (TXP) and handoveroffset Input Control Parameters (ICPs) with every other xApp; the lowest severity pair (MLB–ES) scores 0.018. This pattern is consistent with the framework’s design goal of reserving intervention for episodes where the cost is justified. Many detected conflicts are neither clearly harmless nor clearly severe enough to justify immediate action. Deferral provides a useful intermediate state that allows the controller to accumulate evidence rather than spending mitigation cost prematurely, while bounded deferral (Dmax = 2) prevents indefinite inaction. The confidence intervals in Fig. 2 describe variation across seeds under the fixed scenario; they do not establish robustness across operating conditions. B. Utility of Cost-Aware Triage The second result is that selected interventions remain beneficial on average. Over 30 runs, the mean mitigation benefit is 0.192 (CI [0.171, 0.213]), the mean residual loss is 0.076 (CI [0.069, 0.083]), and the mean net utility per mitigated episode is 0.132 (CI [0.113, 0.151]). The benefitversus-cost view in Fig. 3 shows that the mean operating point lies clearly above the benefit-equals-cost line, confirming worthwhile interventions on average. Fig. 4 shows the mechanism selection rates. qacm is the dominant mechanism at 60.8 % of all selections (CI [55.5, 66.2]), followed by dampen_latest at 14.1 % (CI [9.1, 19.1]) and eg at 9.6 % (CI [7.8, 11.4]). qacm dominates because the episodes that clear the τh threshold are

Benefit vs cost per run (n=30, 95% CI)

TABLE III R ETROACTIVE BASELINE COMPARISON

predominantly direct conflicts, for which qacm carries the highest efficacy (η = 0.88) combined with the lowest stability and opportunity cost coefficients. The unilateral cancel and dampen mechanisms carry higher stability and opportunity costs, making them inferior in net utility for most directconflict episodes under the calibrated model. The game-theoretic satisfaction rate is high at 93.3 % (CI [91.2, 95.3]), confirming that the compromise-oriented mechanisms produce outcomes aligned with the intended fairness objective. Comparison with baseline strategies. Table III compares the proposed framework retroactively against three alternative decision rules applied to the same pre-computed episode utility scores from the 30 runs. Always-tolerate never intervenes and achieves zero per-episode utility. Always-mitigate applies the best mechanism to every episode: it achieves slightly higher per-episode utility (0.034) but requires 5.1× more interventions (118.3 vs 23.4 per run). Positive-utility rule mitigates whenever U ∗ (e) > 0: it achieves the highest per-episode utility (0.047) but ignores uncertainty, acting on 44.8 % of episodes. The proposed framework extracts 0.132 utility per intervention, 3.9× more than always-mitigate, by confining interventions to episodes that clearly exceed τh with low uncertainty. The 0.020 per-episode gap relative to the positive-utility rule reflects the combined effects of the utility threshold, uncertainty gate, and bounded deferral; this comparison does not isolate their individual contributions. Both utility columns use the same U (e, m): per-episode utility divides the sum of intervention utilities by all detected episodes, while per-intervention utility divides it by interventions only. Non-intervention contributes zero incremental mitigation utility, not zero unmanaged loss. Table entries are run-level averages, so multiplying averaged rates and utilities need not reproduce the averaged per-episode score. Selectivity improves utility per intervention, while the positive-utility rule has higher utility per episode. No paired significance test is reported, and these model-based comparisons do not establish superiority over independently implemented PACIFISTA, QACM, or learning-based controllers. C. Interpretability of the Triage Decisions The third result is that the triage outcomes are interpretable. Fig. 5 shows clear differences between the three action groups. Episodes that are finally mitigated have the highest average severity prior, the largest degradation magnitude, and the strongest harmful recurrence. In contrast, tolerated episodes are associated with lower degradation and weaker harmful recurrence, while also showing the highest uncertainty.

0.3

0.25

Avg estimated benefit B

Per-Int utility Nint — 0 0.132 23.4 0.034 118.3 0.103 53.0

0.2

0.15

0.1

0.05

0

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

Avg mitigation cost C

mit

Fig. 3. Average estimated benefit vs. mitigation cost per run. All runs sit above the benefit-equals-cost line; the mean operating point (red square) confirms positive net utility across runs. 70

60

Win rate within mitigations (%)

Policy Per-Ep utility Always-tolerate 0.000 Proposed framework 0.027 Always-mitigate 0.034 Positive-utility rule 0.047

Individual runs Mean ' 95% CI (n=30) Benefit = Cost

50

40

30

20

10

0

st late cel_ can

st rride _late ove pen rity_ dam prio

f nsw

eg

m qac

Fig. 4. Mitigation mechanism selection rates.

These trends match the intended logic, and align with the design goal of limiting interventions to episodes where the cost is clearly justified. Critically, the deferred group sits between the other two on every feature: higher severity and degradation than tolerated, but lower certainty than mitigated. This ordering is consistent with the decision criteria; it does not independently validate the benefit of deferral. D. Discussion The triage setting is conservative: defer dominates at 66.8 %. Unlike ignoring conflicts outright, deferred episodes remain under active observation, and bounded deferral guarantees escalation once Dmax = 2 is reached. This mechanism is consequential: 47.6 % (CI [38.3, 57.0]) of all mitigations across the 30 runs were forced escalations rather than utilitytriggered (U ∗ (e) > τh ). Forced mitigations carry lower mean net utility (+0.042, CI [0.020, 0.064]) than utility-triggered ones (+0.210, CI [0.194, 0.227]), showing that Dmax trades some efficiency for a guarantee that persistent harmful conflicts are not indefinitely postponed by uncertainty; operators may raise Dmax for fewer forced interventions or lower it for promptness. These escalation statistics describe the safety–efficiency trade-off under the calibrated model. A twostate ablation and closed-loop re-observation are needed to determine whether deferral reduces unnecessary interventions or improves long-term network stability.

Action-conditioned feature summary (n=30 runs, 95% CI) Severity prior

0.2

Degradation magnitude

0.8

R EFERENCES

0.7 0.15

0.6 0.5

0.1

0.4 0.3

0.05

0.2 0.1

0

tolerate

defer

mitigate

Uncertainty

0.5

0

0.4

0.8

0.3

0.6

0.2

0.4

0.1

0.2

0

tolerate

defer

tolerate

mitigate

0

defer

mitigate

Harmful recurrence

1

tolerate

defer

mitigate

Fig. 5. Action-conditioned feature distributions.

Parameter dependence and limitations. The thresholds and coefficients encode operator risk preferences. For fixed episode scores, increasing τh or decreasing uh can only shrink the immediate-mitigation set before forced escalation. Coefficient sensitivity follows directly from U (e, m) = ηm,ce Lconf (e) − ξm ue − Cmit (e, m): with other terms fixed, a loss perturbation ∆L changes a candidate’s utility by ηm,ce ∆L, while an added cost ∆C lowers it by ∆C. Decisions near either threshold or near a tie between mechanisms are therefore most sensitive. These are analytical dependencies; action-rate and utility robustness under parameter perturbations have not been measured. A stale severity prior or noisy degradation measurement can cause false tolerance or unnecessary mitigation. Dmax bounds repeated deferral but does not correct false tolerance or guarantee positive utility for forced interventions. Evaluation covers four xApps, one topology, and fixed traffic/mobility settings; transfer to other vendors, conflict types, or loads remains untested. The utility-based results assess the scoring policy’s internal consistency. Independent throughput, latency, energy, and signalling measurements after executing decisions are needed for closed-loop validation. VI. C ONCLUSION This paper presented a cost-aware conflict triage framework that fills a gap in Open RAN conflict management: deciding whether intervention is worth its operational cost, not just how to intervene. By combining unmanaged loss estimation, permechanism cost scoring, and bounded deferral, the framework reserves mitigation for episodes with high estimated benefit relative to cost, deferring borderline or uncertain cases instead of acting prematurely. Over 30 Monte Carlo trials with four concurrent xApps, this yields internally consistent, selective decisions: 0.132 net utility per intervention (3.9× an alwaysmitigate policy) from 23.4 mean interventions per run, with qacm winning 60.8 % of selections and bounded deferral (Dmax = 2) accounting for 47.6 % of mitigations as a safety net under uncertainty. Future work should add closed-loop KPI re-simulation, threshold sensitivity analysis, and validation on a live O-RAN testbed.

[1] A. Wadud, F. Golpayegani, and N. Afraz, “xApp-Level Conflict Mitigation in O-RAN, a Mobility Driven Energy Saving Case,” in IEEE INFOCOM 2025-IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS). IEEE, 2025, pp. 1–6. [2] C. Adamczyk and A. Kliks, “Conflict Mitigation Framework and Conflict Detection in O-RAN Near-RT RIC,” IEEE Communications Magazine, vol. 61, no. 12, pp. 199–205, 2023. [3] P. Brach del Prever, S. D’Oro, L. Bonati, M. Polese, M. Tsampazi, H. Lehmann, and T. Melodia, “PACIFISTA: Conflict Evaluation and Management in Open RAN,” IEEE Transactions on Mobile Computing, vol. 24, no. 10, pp. 10 590–10 605, 2025. [4] A. Wadud, F. Golpayegani, and N. Afraz, “QACM: QoS-Aware xApp Conflict Mitigation in Open RAN,” IEEE Transactions on Green Communications and Networking, vol. 8, no. 3, pp. 978–993, 2024. [5] ——, “Conflict Management in the Near-RT-RIC of Open RAN: A Game Theoretic Approach,” in 2023 IEEE Green Computing & Communications (GreenCom). IEEE, 2023, pp. 479–486. [6] I. Cinemre, T. Mahmoodi, and A. Farzaneh, “xapp conflict mitigation with scheduler,” arXiv preprint arXiv:2504.06867, 2025. [7] A. Giannopoulos, S. Spantideas, L. George, K. Alexandros, and P. Trakadas, “COMIX: Generalized Conflict Management in O-RAN xApps – Architecture, Workflow, and a Power Control Case,” arXiv preprint arXiv:2501.14619, 2025. [8] K. Slimani, C. F. Da Silva, L. Médini, and P. Ghodous, “Conflict mitigation in collaborative design,” International Journal of Production Research, vol. 44, no. 09, pp. 1681–1702, 2006. [9] M. Katsigiannis and T. Smura, “A cost model for radio access data networks,” info, vol. 17, no. 1, pp. 39–53, 2015. [10] A. Wadud, F. Golpayegani, and N. Afraz, “Ai-driven multi-modal adaptive handover control optimization for o-ran,” arXiv preprint arXiv:2603.17158, 2026. [11] H. Wang and B. Li, “Double-deep q-learning-based handover management in mmwave heterogeneous networks with dual connectivity,” Transactions on Emerging Telecommunications Technologies, vol. 35, no. 1, p. e4907, 2024. [12] H. Zafar, E. Tohidi, M. Kasparick, and S. Stańczak, “Conflict mitigation approach for o-ran xapps,” in 2025 IEEE Wireless Communications and Networking Conference (WCNC). IEEE, 2025, pp. 1–6. [13] M. Hassan, A. Diab, S. Parameswaran, and A. Mitschele-Thiel, “Machine learning-based coverage and capacity optimization xapp/rapp for open ran 5g campus networks,” in Mobilkommunikation; 28. ITGFachtagung. VDE, 2024, pp. 191–196.

Related documents

• Cost-Aware Conflict Triage for xApp Control Conflicts in Open RAN#1326560
• Software Quality Assurance#1334347
• Contabilidade de seguros#1239788
• Software architecture#1190512
• Intellectual Property, Computer Software and the Open Source Movement#990231