Quality-preserving Model for Electronics Production Quality Tests Reduction

Noufa Haneefa; Teddy Lazebnik; Einav Peretz-Andersson

cs.LGApr 7, 2026

Quality-preserving Model for Electronics Production Quality Tests Reduction

Authors: Noufa Haneefa, Teddy Lazebnik, Einav Peretz-Andersson

Organizations: Department of Computing, Jonkoping University, Jonkoping, Sweden · Department of Information Science, University of Haifa, Haifa, Israel

Abstract

Manufacturing test flows in high-volume electronics production are typically fixed during product development and executed unchanged on every unit, even as failure patterns and process conditions evolve. This protects quality, but it also imposes unnecessary test cost, while existing data-driven methods mostly optimize static test subsets and neither adapt online to changing defect distributions nor explicitly control escape risk. In this study, we present an adaptive test-selection framework that combines offline minimum-cost diagnostic subset construction using greedy set cover with an online Thompson-sampling multi-armed bandit that switches between full and reduced test plans using a rolling process-stability signal. We evaluate the framework on two printed circuit board assembly stages-Functional Circuit Test and End-of-Line test-covering 28,000 board runs. Offline analysis identified zero-escape reduced plans that cut test time by 18.78% in Functional Circuit Test and 91.57% in End-of-Line testing. Under temporal validation with real concept drift, static reduction produced 110 escaped defects in Functional Circuit Test and 8 in End-of-Line, whereas the adaptive policy reduced escapes to zero by reverting to fuller coverage when instability emerged in practice. These results show that online learning can preserve manufacturing quality while reducing test burden, offering a practical route to adaptive test planning across production domains, and offering both economic and logistics improvement for companies.

Figures & tables

Figure 2: A schematic view of the multi-arm bandit algorithms and their production tests prediction.

Symbol	Definition
$\mathcal{T}=\{1,\ldots,M\}$	Index set of all available test steps
$M$	Total number of test steps
$c_{i}$	Execution cost (mean time in seconds) of test step $i$
$c_{\text{full}}$	Total cost of executing all test steps
$N$	Number of historical units in training dataset
$\mathbf{Y}\in\{0,1\}^{N\times M}$	Binary outcome matrix

Table 1: Summary of the parameters used in the model with their notation and definition.

Figure 3: Schematic flow of the proposed framework.

Symbol	Equation	Explanation
$C_{\text{red}}$ selection rate ( $\mathrm{SelRate}(\pi)$ )	$\displaystyle\frac{1}{N}\sum_{t=1}^{N}\mathbf{1}[a_{t}=1]$	Measures the fraction of boards for which the MAB agent selected the reduced test plan rather than the full plan. Here, $a_{t}\in\{0,1\}$ is the arm selected for board $t$ , and $N$ is the total number of boards processed. A value of $1.0$ means the agent always selected the reduced plan, while a value of $0$ means the full plan was always used. Higher values indicate greater exploitation of the cost-saving opportunity.
Cost saving (%) ( $\mathrm{Saving}(\pi)$ )	$\displaystyle\left(1-\frac{\bar{c}_{\pi}}{c_{\text{full}}}\right)\times 100,\qquad\bar{c}_{\pi}=\frac{1}{N}\sum_{t=1}^{N}c_{a_{t}}$	Measures the percentage reduction in mean test time per board relative to executing the full sequence on every board. Here, $c_{a_{t}}$ is the cost of the plan selected for board $t$ , and $c_{\text{full}}=\sum_{i\in\mathcal{T}}c_{i}$ is the cost of the full test sequence. A saving of $0\%$ corresponds to always running the full plan, while the maximum achievable saving corresponds to always running $C_{\text{red}}$ .
Escaped units ( $\mathrm{Escaped}(\pi)$ )	$\displaystyle\sum_{u\in U_{F}}\mathbf{1}\{D_{\pi(y_{u})}(u)=0\}$	Counts the absolute number of genuinely defective boards that were not detected by the executed test plan and would therefore have proceeded to shipment undetected. Here, $D_{C}(u)=\max_{i\in C}(1-y_{u,i})$ . A value of zero means all defective boards were caught. Any non-zero value represents a direct quality failure: a defective board that reached the customer undetected.
Empirical escape risk ( $\hat{R}(\pi)$ )	$\displaystyle\frac{1}{\|U_{F}\|}\sum_{u\in U_{F}}\mathbf{1}\{D_{\pi(y_{u})}(u)=0\}$	Expresses the same quantity as a fraction of the total defect population, making it comparable across datasets with different numbers of failing boards. A value of $\hat{R}(\pi)=0$ indicates complete defect coverage, meaning that all defective boards were detected. This metric is directly comparable to the quality threshold $\delta$ , which requires $\hat{R}(\pi)\leq\delta$ at all times.

Table 2: Evaluation metrics used to assess the policy

\pi

Test	Cohort	Algorithm	$\beta$	$C_{\text{red}}$ (%)	Saving (%)
FCT	Training	Baseline ( $C_{\text{full}}$ )	—	0.0	0.00
		Static $C_{\text{red}}$	—	100.0	18.78
		Epsilon-Greedy	10	89.6	16.83
		UCB	10	92.8	17.42
		Thompson Sampling	10	93.8	17.61
	Validation	Baseline ( $C_{\text{full}}$ )	—	0.0	0.00

Table 3: Comparative results on the FCT and EOL datasets across training and validation cohorts.

\beta

is the instability sensitivity parameter, and

\hat{R}(\pi)

denotes empirical escape risk.

Panel of Figure 4 — Figure 4: Cost-risk Pareto fronts for the two test suites. The grey dotted line shows a logarithmic fit to the Pareto-optimal points, and the red star marks the zero-escape operating point.

Panel of Figure 5 — Figure 5: Rolling $C_{\text{red}}$ selection fraction during training for the Thompson Sampling agent on the FCT and EOL datasets.

Panel of Figure 6 — Figure 6: Rolling pass rate and $C_{\text{red}}$ selection behavior during validation for the FCT and EOL datasets.

Manufacturing

Manufacturing research develops methods and systems to design, plan, automate, monitor, and improve industrial production.

See all 77 papers

Adaptive Sequential Test Planning for Multi-Mechanism Reliability Qualification via Bayesian Monte Carlo Tree Search
Aug 10, 2026Youssef A. Elhagrasy, Ian Hill, André IvanovSemiconductor Manufacturing Monte Carlo Tree Search
Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC V6T 1Z4, Canada
A probabilistic framework for online test-time adaptation
Jun 24, 2026Daniel Corrales, David Ríos InsuaDistribution Shift Robustness Test-Time Adaptation
Escuela de Doctorado, Universidad Aut´onoma de Madrid, 28049 Madrid, Spain · Institute of Mathematical Sciences, ICMAT-CSIC, 28049 Madrid, Spain
Epistemic Uncertainty-Aware Defect Detection for Quality Control in Medical Device Manufacturing
Oct 6, 2026Raham A. Butt, Marco Romanelli, Roche C. de GuzmanSelective Prediction Epistemic Uncertainty
Department of Engineering, Hofstra University, Hempstead, NY 11549 USA · Department of Computer Science, Hofstra University, Hempstead, NY 11549 USA

Aug 10, 2026J/K move · Enter open · S save

Adaptive Sequential Test Planning for Multi-Mechanism Reliability Qualification via Bayesian Monte Carlo Tree Search

Youssef A. Elhagrasy, Ian Hill, André Ivanov

Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC V6T 1Z4, Canada

Semiconductor Manufacturing Monte Carlo Tree Search Bayesian Experimental Design

Open paper AskPDF arXiv

Reliability qualification of advanced semiconductor devices requires sequential stress decisions that balance characterization objectives against multiple competing failure mechanisms. Current practice relies on static test plans derived from population-level acceleration models, which cannot adapt to per-unit variability or real-time degradation observations. This paper presents a closed-loop adaptive test planning framework that formulates reliability qualification as a partially observable sequential decision problem and solves it using Monte Carlo tree search for seed-action simulators (MCTS-SA) coupled with extended Kalman filter (EKF) belief-state estimation. The framework models stochastic, per-device variability in bias temperature instability (BTI), electromigration (EM), and time-dependent dielectric breakdown (TDDB), and treats stress selection as a constrained sequential optimization, i.e., to maximize the probability of successful degradation characterization while respecting catastrophic failure constraints. Under the experimental assumptions used here (discrete stress actions, proxy damage observability, and cumulative degradation without recovery), we believe this to be a novel application of tree-search-based adaptive test planning to multi-mechanism reliability qualification. Across 5,000 planning iterations, the characterization yield (CY) improves from 20% in the first 500 iterations to over 54% in the final 500, with 39% cumulative success, while the best successful test sequence terminates with EM and TDDB damage fractions DEM=0.564 and DTDDB=0.537, well within safety margins. These results demonstrate that sequential Bayesian planning can synthesize damage-aware test policies that significantly outperform non-adaptive strategies for reliability qualification under competing failure modes.

Abstract

Manufacturing

Explore similar work

Adaptive Sequential Test Planning for Multi-Mechanism Reliability Qualification via Bayesian Monte Carlo Tree Search

A probabilistic framework for online test-time adaptation

Epistemic Uncertainty-Aware Defect Detection for Quality Control in Medical Device Manufacturing