Production and supply chain optimization problems have grown increasingly complex due to advancements in computing power and mixed-integer linear programming (MILP) solvers. Algebraic modeling languages (AMLs) such as Pyomo, JuMP, AMPL and GAMS, enable researchers and practitioners to describe these problems in a high-level, declarative fashion in a given programming language. This paper presents a comparative analysis of these AMLs with an emphasis on their use in real-world problems. A conceptual framework that empowers traditional modeling languages is presented as a potential graph-based MILP modeling platform. It enables the creation of sets, variables, constraints and objectives as nodes on graph. A visual equation editor is proposed to increase the visibility, explainability and the modeling process. The framework also provides a solver and AML agnostic interfaces by allowing users to seamlessly switch between different solving environments. The article concludes with a discussion of further research directions.