Purpose: The purpose of this paper is to develop an Internet of Things (IoT)-based framework for supply chain (SC) performance measurement and real-time decision alignment. The aims of the proposed model are to optimize the performance indicator based on integrated supply chain operations reference metrics.Design/methodology/approach: The SC multi-dimensional structure is modeled by multi-objective optimization methods. The operational presented model considers important SC features thoroughly such as multi-echelons, several suppliers, several manufacturers and several products during multiple periods. A multi-objective mathematical programming model is then developed to yield the operational decisions with Pareto efficient performance values and solved using a well-known meta-heuristic algorithm, i.e., non-dominated sorting genetic algorithm II. Afterward, Technique for Order of Preference by Similarity to Ideal Solution method is used to determine the best operational solution based on the strategic decision maker’s idea.Findings: This paper proposes a dynamic integrated solution for three main problems: strategic decisions in high level, operational decisions in low level and alignment of these two decision levels.Originality/value: The authors propose a human intelligence-based process for high level decision and machine intelligence-based decision support systems for low level decision using a novel approach. High level and low level decisions are aligned by a machine intelligence model as well. The presented framework is based on change detection, event driven planning and real-time decision alignment.

In supply chain optimisation problems, determining the location, number and capacity of facilities is concerned as strategic decisions, while mid-term and short-term decisions such as assembly policy, inventory levels and scheduling are considered as the tactical and operational decision levels. This paper addresses the optimisation of strategic and tactical decisions in the supply chain network design (SCND) under demand uncertainty. In this respect, a two-stage stochastic programming model is developed in which strategic location decisions are made in the first-stage, while the second-stage contains SCND problem and the assembly line balancing as a tactical decision. In the solution scheme, the combination of sample average approximation and Latin hypercube sampling methods is utilised to solve the developed two-stage mixed-integer stochastic programming model. Finally, computational experiments on randomly generated problem instances are presented to demonstrate the performance and power of developed model in handling uncertainty. Computational experiments showed that stochastic model yields better results compared with deterministic model in terms of objective function value, i.e. the sum of the first-stage costs and the expected second-stage costs. This issue proved that uncertainty would be a significant and fundamental element of developed model and improve the quality of solutions.