Koopman Mode Analysis of agent-based models of logistics processes.

摘要: Modern logistics processes and systems can feature extremely complicated dynamics. Agent Based Modeling is emerging as a powerful modeling tool for design, analysis control of such systems. However, the complexity model itself be overwhelming mathematical meta-modeling tools are needed that aggregate information enable fast accurate decision making system design. Here we present Koopman Mode Analysis (KMA) tool. KMA uncovers exponentially growing, decaying or oscillating collective patterns in dynamical data. We apply methodology to two problems, both which exhibit bifurcation behavior, but very different dynamics: Medical Treatment Facility (MTF) ship fueling (SF) logistics. The MTF problem features transition between efficient operation at low casualty rates inefficient beyond critical rate, while SF short mission life initial fuel levels sustained level. Both bifurcations detected by analyzing spectrum associated operator. Mathematical provided justifying use Dynamic Decomposition algorithm punctuated linear decay dynamics featured problem.