Dynamic Windows Scheduling with Reallocation

We consider the Windows Scheduling (WS) problem, which is a restricted version of Unit-Fractions Bin Packing, and it also called Inventory Replenishment in context Supply Chain. In brief, WS problem to schedule use communication channels clients. Each client c i characterized by an active cycle window w . During period time that any given active, there must be at least one transmission from scheduled consecutive slots, but most can carried out each channel per slot. The goal minimize number used. extend previous online models, where decisions are permanent, assuming clients may reallocated some cost. assume such cost constant amount paid reallocation. That is, we aim reallocations. present three reallocation algorithms for Scheduling. evaluate experimentally multiple variants these protocols showing that, practice, all achieve amortized reallocations with close optimal usage. Our simulations expose interesting tradeoffs between introduce new objective function applied models not possible. analyze this metric best our knowledge, first protocol theoretical guarantees applies scenarios leave analysis against current load rather than peak load. Using results, observe bounds on usage algorithms.