The stability requirement for node enterprises within Ecological Industrial Chain (EIC) has a high degree of functional similarity with the reliability of engineering components. Thus, a methodology for EIC design optimization is established in this study based on the reliability theory. Results showed that series structure had weaker stability than parallel structure. The stability of entire system is reduced with the increase of the number of series, and node enterprises with smaller reliability showed greater contribution to the system stability. On contrary, for the parallel structure, node enterprises with higher reliability had greater contribution to the system stability. To improve the system stability, it is crucial to maintain the weak parts of system or to introduce an appropriate number of “spare parts”, i.e. to adopt parallel structure and reduce the number of enterprises series. For the parallel structure, priority must be given to those enterprises with higher reliability. Similarly, other constraints (e.g. economic costs and resource constraints) should also be taken into consideration. The redundant design of EIC becomes nonlinear planning decision problems. Novel nonlinear programming model for redundant EIC design is proposed in this study based on the reliability theory, and is effectively verified by means of MATLAB software programming and empirical analysis. Results provide a theoretical basis and practical reference for decision making process of optimizing the design and management of EIC.