This systematic literature review synthesises recent research on MCO of fuel mixtures and engine performance for biofuel and alternative-fuel applications. Paper surveys deterministic techniques (e.g. Taguchi, polynomial regression) and heuristic/evolutionary methods (e.g. ANN, GA, NSGA-II/III, AGE-MOEA), and assess their application to a range of fuels including biodiesel, plastic pyrolysis oil (PPO), hydrogen blends and nanoparticle-enhanced fuels (Al2O3, CeO2, CNT). Reviewed studies commonly optimise blend ratio, injection timing and pressure, compression ratio, load/speed and additive concentration, reporting typical improvements in brake thermal efficiency (BTE) and reductions in specific fuel consumption (SFC) and CO/HC emissions (with BTE gains often cited around 30–35% under laboratory optimal conditions). However, methods differ in scope: RSM/Taguchi deliver efficient, interpretable local models while ANN and evolutionary algorithms enable Pareto-front exploration but demand larger datasets and validation. Major gaps include limited long-term durability testing, scarce techno-economic and life-cycle (LCA) assessments, insufficient cross-engine validation, and weak reproducibility due to incomplete reporting of raw data and statistical metrics. The recommendations include for wider adoption of Pareto-based multi-objective frameworks, integration of economic and LCA criteria, explicit uncertainty analysis, standardised reporting of factor ranges and validation statistics, and open data practices to accelerate translation of optimised blends from bench to practice.
Malashin et al. (Mon,) studied this question.