The design of efficient search algorithms is central to solving combinatorial optimization problems (COPs). Such problems often involve equality and/or inequality constraints, which make them more challenging. A common strategy to address these issues is the penalty function approach, where constrained problems are reformulated as unconstrained ones (uCOPs). The Binary Artificial Electric Field Algorithm (B-AEFA) is a recent extension of the original AEFA, adapted to binary space to enhance its effectiveness in discrete domains. In B-AEFA, transfer functions are employed to map continuous values into binary ones, a technique frequently adopted to improve the exploration and exploitation balance in search processes. In this work, B-AEFA is applied to a well-known COP, namely the 0–1 knapsack problem, across both small- and large-scale instances. Constraint handling is achieved using the Augmented Lagrangian Method (ALM), which has proven effective for such formulations. B-AEFA is benchmarked against six state-of-the-art binary optimization algorithms, including a hybrid scheme, and its performance is further assessed using a non-parametric statistical test. The results indicate that B-AEFA consistently delivers superior outcomes compared to the competing methods, highlighting its capability to solve constrained combinatorial optimization problems efficiently.