arXiv subjects

Gerald M. Knapp

Publications and source records attributed to Gerald M. Knapp.

5 recordsLinked to original sources

arXiv2026

Dynamic Multi-Depot Vehicle Routing with Online Requests: Event-Driven Transformer--DRL and Rolling-Horizon Benchmarking

This paper presents an event-driven learning and benchmarking framework for the Dynamic Multi-Depot Vehicle Routing Problem with progressively revealed requests and evolving vehicle states. Masked MLP and Transformer policies are trained through behavior cloning and proximal policy optimization. Deterministic feasibility masking prevents invalid vehicle--request assignments, while fixed-prefix/flexible-suffix route commitments protect completed, active, and near-term decisions and separately measure vehicle reassignment and resequencing. The learned policies are compared with dynamic insertion heuristics and time-limited rolling-horizon optimization. In a 20-scenario policy benchmark, all methods completed every request without invalid actions, but nearest feasible achieved the lowest mean objective and outperformed the learned policies in routing quality, waiting time, stability, makespan, and runtime. Across five independent training runs, PPO had little average effect on the MLP and improved the Transformer on average, although with greater seed variability. Under the common protocol, nearest feasible achieved the lowest combined objective and route disruption, whereas rolling horizon achieved the lowest waiting times and makespan at substantially higher computational cost. The learned policies retained millisecond-level decisions and transferred to instances with up to 80 requests without retraining, but did not outperform the strongest heuristic. No single method was best across routing efficiency, service responsiveness, stability, and online computation.

cs.LG↗

arXiv2026

A Graph Neural Network--Guided Genetic Algorithm for Physical Internet Supply Chain Optimization under Cost Uncertainty

Inventory and distribution planning in Physical Internet networks requires coordinating factory-hub assignments, factory supply, lateral transshipment among collaborative hubs, retailer deliveries, and shortages. The problem combines discrete assignment decisions with interdependent continuous flows, while uncertain operating costs make robust planning more difficult. This study formulates deterministic and min-max regret models for a three-echelon network of factories, hubs, and retailers and develops a graph neural network-guided genetic algorithm (GNN-GA) for the assignment decisions. The GNN estimates hub-specific factory-selection probabilities that are used to construct the initial GA population and adapt mutation according to prediction uncertainty. Each previously unseen candidate assignment is evaluated by solving the remaining continuous-flow problem to LP optimality. Simulated annealing, a standard GA, and GNN-GA are compared on 15 instances using matched random seeds and fixed limits on distinct assignment evaluations. Because the evaluation budgets for test Instances 13-15 are smaller than the nominal population size, these experiments primarily assess the quality of learned initialization rather than multi-generation evolutionary search. A separate 400-evaluation experiment on exact test Instance 13 permits three complete offspring generations and a partial fourth pass, with GNN-GA outperforming GA in all 10 matched runs. Three independently generated exact-solvable instances provide a separate test of transfer. Ablation results show that learned initialization provides most of the improvement, while entropy-guided mutation has a smaller, instance-dependent effect. Per-instance solution times include GNN inference and search but exclude model training and one-time model setup.

cs.NE↗

arXiv2026

A Deep Reinforcement Learning (DRL)-Based Transformer Method for Solving the Open Shop Scheduling Problem

The open shop scheduling problem (OSSP) arises in many industrial and service settings but remains computationally challenging as the number of jobs and machines increases. While exact methods quickly become intractable, classical dispatching rules and metaheuristics may require substantial tuning to maintain solution quality at large scales. This study develops a Transformer-based scheduling policy for OSSP using an encoder-decoder architecture with multi-head attention. The model is trained on Taillard benchmark instances (4x4, 5x5, 7x7, and 10x10) using only the processing-time matrix as input and produces feasible schedules with makespans typically within 15-30% of best-known values. To evaluate scalability, the trained policy is applied without retraining to randomly generated instances from 40x40 to 100x100 and compared against classical dispatching heuristics, including SPT, LPT, MWKR, and EST. Across these large instances, the Transformer achieved average gaps of 12.89-15.12% relative to a standard lower bound. Compared with EST, the Transformer remained competitive, typically within a modest margin, while substantially outperforming SPT and LPT. These results indicate that a Transformer policy trained on small OSSP instances can generalize to substantially larger problems and provide a feature-light, learning-based alternative to classical dispatching rules.

cs.AI↗

arXiv2019

Improving Suppression to Reduce Disclosure Risk and Enhance Data Utility

In Privacy Preserving Data Publishing, various privacy models have been developed for employing anonymization operations on sensitive individual level datasets, in order to publish the data for public access while preserving the privacy of individuals in the dataset. However, there is always a trade-off between preserving privacy and data utility; the more changes we make on the confidential dataset to reduce disclosure risk, the more information the data loses and the less data utility it preserves. The optimum privacy technique is the one that results in a dataset with minimum disclosure risk and maximum data utility. In this paper, we propose an improved suppression method, which reduces the disclosure risk and enhances the data utility by targeting the highest risk records and keeping other records intact. We have shown the effectiveness of our approach through an experiment on a real-world confidential dataset.

cs.DB↗

arXiv2019

A Novel Microdata Privacy Disclosure Risk Measure

A tremendous amount of individual-level data is generated each day, of use to marketing, decision makers, and machine learning applications. This data often contain private and sensitive information about individuals, which can be disclosed by adversaries. An adversary can recognize the underlying individual's identity for a data record by looking at the values of quasi-identifier attributes, known as identity disclosure, or can uncover sensitive information about an individual through attribute disclosure. In Statistical Disclosure Control, multiple disclosure risk measures have been proposed. These share two drawbacks: they do not consider identity and attribute disclosure concurrently in the risk measure, and they make restrictive assumptions on an adversary's knowledge by assuming certain attributes are quasi-identifiers and there is a clear boundary between quasi-identifiers and sensitive information. In this paper, we present a novel disclosure risk measure that addresses these limitations, by presenting a single combined metric of identity and attribute disclosure risk, and providing flexibility in modeling adversary's knowledge. We have developed an efficient algorithm for computing the proposed risk measure and evaluated the feasibility and performance of our approach on a real-world data set from the domain of social work.

cs.CR↗