An objective function assigns a quantitative value to each feasible decision and defines what an optimization model maximizes or minimizes. In linear programming, it combines decision variables linearly, subject to constraints representing limits such as budget, resources, or strategic requirements. section 8.0 section 8.2
For feature prioritization, let xᵢ indicate whether feature i is selected and cᵢ represent its priority score. The objective is to maximize Σ cᵢxᵢ. With binary variables, xᵢ = 1 selects a feature and xᵢ = 0 excludes it. The model can also minimize costs or optimize another quantitative organizational goal. section 8.2 section 8.3
Maximizing feature value per effort
For each feature, calculate priority as value_points / effort_points. Because scipy.optimize.linprog minimizes by default, negate each score: c = [-f.value_points / f.effort_points for f in features]. Solving with budget limits and three-to-four-feature bounds per epic then minimizes the negative total, which is equivalent to maximizing total priority. section 8.2.2 section 8.2.3
>>> c = [-f.value_points / f.effort_points for f in features]
Shows the compact Python expression that constructs objective coefficients from feature value and effort points. section 8.2.3
The objective function does not determine feasibility by itself. Linear constraints must restrict the same decision variables, and the resulting solution should be checked for solver success, selected features, costs, priority scores, and compliance with organizational limits. section 8.2.2 section 8.3
