Customer lifecycle decisions in enterprises are often fragmented across marketing, finance, customer experience, and operations, resulting in inconsistent actions and suboptimal outcomes. This paper introduces Multi-Agent Decision Intelligence (MADI), a framework that models enterprise functions as autonomous yet coordinated agents aligned through shared objectives. Using centralized training with decentralized execution, agents negotiate actions that balance lifetime value growth, churn reduction, customer experience, and cost-to-serve efficiency under operational and regulatory constraints. Experimental results on industrial and semi-synthetic datasets demonstrate consistent improvements over siloed optimization, centralized reinforcement learning, and heuristic baselines. A pilot deployment within a Saudi enterprise further confirms practical feasibility, improved crossfunctional alignment, and reduced decision latency. MADI provides an enterprise-ready blueprint for coordinated AI-driven decisionmaking in complex organizational environments.
Practical knowledge on the design and application of multi-objective evolutionary algorithms (MOEAs) is available but well-founded theoretical analyses of the runtime are rare. Laumanns, Thiele, Zitzler, Welzel and Deb (2002) have started such an analysis for two simple mutation-based algorithms including SEMO. These algorithms search locally in the neighborhood of their current population by selecting an individual and flipping one randomly chosen bit. Due to its local search operator, SEMO cannot escape from local optima, and, therefore, has no finite expected runtime in general. In this talk, we investigate the runtime of a variant of SEMO whose mutation operator flips each bit independently. It is proven that its expected runtime is O(n^n) for all objective functions f: {0,1}^n -> R^m, and that there are bicriteria problems among the hardest problem for this algorithm. Moreover, for each d between 2 and n, a bicriteria problem with expected runtime Theta(n^d) is presented. This shows that bicriteria problems cover the full range of potential runtimes of this variant of SEMO. For the problem LOTZ (Leading-Ones-Trailing Zeroes), the runtime does not increase substantially if we use the global search operator. Finally, we consider the problem MOCO (Multi-Objective-Counting-Ones). We show that the conjectured bound O((n^2)log n) on the expected runtime is wrong for both variants of SEMO. In fact, MOCO is almost a worst case example for SEMO if we consider the expected runtime; however, the runtime is O((n^2)log n) with high probability. Some ideas from the proof will be presented.