针对第四次工业革命与大数据信息时代对核心智能新技术、新方法的迫切需求，首先，我们引进新的归一化的由大数据流驱使的“机器侧-物理侧”基于多种类区块链智能服务的混合网络交互系统，它包含通信网、云中心、能源电力、金融科技系统及其混合系统作为特例，这种双侧系统是当前世界上所关注物联网的最新一般形式；其次，我们用一般性三随机更新报酬过程或由Levy过程驱使的复合过程来精准刻画当前国际上最为流行的三维度大数据流的概念，其三维统计特性由瞬时大容量数据批量到达、大时变随机输入速率及高维多样数据流展现，在此假设前提与一般性效用等准则函数意义下，并在包含无限缓存、有限缓存、零缓存及其混合的各种复杂情况下，我们将建立起一般性（随机或排队）动力学系统模型及得到可用于系统资源瞬时调度、路由与控制的多目标整体Pareto最优动态博弈均衡策略并发展出归一化的数学理论框架与统一的扩散逼近及随机偏微分决策、论证及求解体系。

To the imperative requirements of new intelligent technologies and methods in the era of Fourth Industrial Revolution and Big Data, we have two folds in our project. Firstly, we establish a unified “Machine Side-Physical Side” system with multiple interactive networks and things. This system is driven by Big Data and operates (or can provide smart services based) on multiple BlockChains. It covers many existing or emerging systems as special cases, e.g., communication networks, cloud-computing and service centers, energy and power grids, FinTech systems, and their mixtures. Secondly, we use the triply stochastic renewal reward processes or Levy driven compound processes to model the popular concept of three-dimensional Big Data in the todays’ world, whose statistical features are characterized by instantaneous high-volume data arrivals, high-speed random and time-varying input rates, and multiple types of data flows. With the triply stochastic processes and the general utility or other criteria functions, we will establish a general (stochastic or queueing) dynamical model for the “Machine Side-Physical Side” system under different buffer regimes: e.g., zero, finite, and infinite ones. Furthermore, based on this model, we will design scheduling, routing, and control policies that are proven to be Pareto optimal Nash equilibrium strategies for stochastic dynamical games with multiple users and objectives. Unified mathematical framework and theory in the proof will be established by establishing diffusion approximation and developing stochastic partial differential equation based decision-making and resolution methods.