This chapter examines a stochastic optimization problem that aims to minimize the total completion time—including both travel and product assembly durations—for a novel type of urban delivery scenario involving the integration of delivery and on-site assembly tasks. Due to various environmental and operational factors such as site conditions and worker availability, the assembly time is subject to uncertainty and is modeled as a random variable. To address this, a chance-constrained programming model is developed, which captures the probabilistic nature of assembly time through probability constraints. Historical data from an enterprise partner is utilized, and statistical learning techniques are applied to estimate the distributional characteristics of the uncertain parameters. The resulting stochastic model is then reformulated into a deterministic equivalent to enhance tractability. A tailored solution algorithm, combining two sub-heuristic strategies, is introduced to efficiently solve the model. Numerical experiments using real-world datasets demonstrate that the proposed method effectively reduces overall completion time, decreases the number of vehicles required, and improves workload balance among them, offering practical value for urban logistics operations.