Antibiotic resistance poses a significant global challenge as bacteria evolve in response to antibiotic use, leading to prolonged hospitalizations, increased healthcare costs, and higher mortality rates. Cephalosporins, a class of beta-lactam antibiotics, are commonly employed to manage infections; however, their misuse and overuse have contributed to resistance development. In response, in silico methods have emerged as cost-effective and efficient tools for drug discovery. This research aims to predict new compounds using ligand-based pharmacophore models while optimizing existing drugs. We employed a de novo approach to synthesize models of cephalosporin structural motifs, integrating the β-lactam core with potential antibiotic candidates. A shared features pharmacophore (SFP) model was constructed using cephalosporins from PubChem, including cephalothin, ceftriaxone, and cefotaxime. The model comprises hydrogen bond acceptors, hydrogen bond donors, aromatic rings, hydrophobic regions, and negatively ionizable sites. Its robustness was evidenced by a goodness-of-hit (GH) score of 0.739. The generated pharmacophore model, with a score of 0.9268, was utilized to screen a drug library, initially assessing 19 compounds. After the drug-likeness screening, seven promising compounds were identified. These candidates were then fused with the cephalosporin core using genetic algorithms and fragment-based design, resulting in 30 novel synthetic models. Most of these models demonstrated a cephalosporin core, over 70 % average similarity, a TPSA (NO) ≤ 99.85 Å², a drug-likeness (QED) ≥ 0.6, and a Synthetic Accessibility Score (SAScore) ≤ 4.3. Molecular docking and MD simulation evaluations highlighted two candidates—Molecule 23 and Molecule 5, demonstrating superior binding affinities to Penicillin-binding protein 1a (PDB ID: 2V2F) compared to controls. To ensure feasible synthesis, molecular architecture comparison and computational retrosynthesis were performed, confirming the likelihood of successful laboratory synthesis. These findings advance the fight against antimicrobial resistance by establishing a method for designing new, highly effective antibiotic drugs.