The use of concrete pavement has significantly increased due to its advantages over asphalt pavement. Determining the concrete mix ratio for asphalt is a vital and crucial step in the construction process. Concrete should be strong, durable, and resistant to environmental degrading factors. This means it needs to withstand freezing, thawing, shrinkage, and harsh environmental conditions (such as heat and cold. These characteristics result in concrete having a longer lifespan and creating robust structures). One of the challenges in design is balancing quality and cost, which comes with its own complexities.. Over the past few years, the application of models and algorithms for multi-objective problems has been a focus of research. This study introduces adaptive neuro-fuzzy inference system goal programming (NFGPM) and fuzzy-goal programming model (FGPM) as expanded variants of basic goal programming models, as alternative tools for allocating asphalt concrete pavement mixtures proportions with multiple, varied objectives. The actual data laboratory experiment datasets were generated and used to develop proposed models. The outcomes of the proposed NFGPM’s mixture proportions and its prediction of concrete properties —such as slump, flexural strength, abrasion resistance, shrinkage and freeze–thaw behavior— are compared with experimental data. The study confirms that the adaptive neuro-fuzzy inference system goal programming model proposed herein can deliver the most cost-effective and optimally performing concrete pavement mixture proportions.