An intelligent zone-based delivery scheduling approach

This paper introduces a zone-based delivery scheduling approach developed using artificial intelligence techniques. In this approach, delivery scheduling is conducted at three different levels: (1) classification of past delivery demand patterns and prediction of future delivery demand using a multi-level pattern clustering and matching method, (2) creation of delivery zones, including their center locations, delivery frequencies, and delivery cost rates, for each of these delivery demand patterns, and (3) identification of the optimal delivery methods, sequence, and timing parameters of delivery tasks. The system was implemented using Smalltalk, an object oriented programming language.