To prevent algorithms from prioritizing revenue over social equity, we must transition from simple profit maximization to multi-objective optimization frameworks. Standard efficiency metrics often focus on throughput or revenue per unit, which inherently favors high-paying industrial users. To counter this, we can integrate a social welfare function into the mathematical definition of efficiency.
One effective method is using the Nash Bargaining Solution or the Social Welfare Function based on proportional fairness. Instead of maximizing total revenue, the objective function maximizes the product of utility gains across all users. This logarithmic approach ensures that the marginal benefit of serving a low-income area is weighted heavily, as a small increase in their utility yields a larger mathematical improvement than a large increase for a wealthy user.
Another approach involves adding equity constraints, such as the Rawlsian Maximin Principle. This mathematical rule dictates that efficiency is achieved only when the utility of the least-advantaged user is maximized. By treating equity as a hard constraint rather than a secondary goal, developers can ensure that the algorithm maintains a baseline level of service for residential areas regardless of their economic output.
