Design and implementation of a rule-based decision support system for dynamic customer relationship management

Today, business customers are quite demanding and they expect on time delivery, short lead times, high quality and affordable prices. In addition, they have different expectations, preferences, and tolerances. On the other side, manufacturing companies have limited resources, and they are confronted with many complex production planning and control (PPC) decisions. In this regard, integrating customer relationship management (CRM) and PPC approaches help companies to build production plans or strategies around the customers, focus on key customers, offer more customized solutions and obtain long term business relationships. This dissertation aims to develop a decision support system (DSS) which integrates CRM and PPC approaches to use manufacturing capabilities more effectively in satisfying customers. To this aim, a job shop system is dealt with and lot streaming is applied to accelerate production flow. In sublot scheduling phase, dynamic scheduling is performed by considering machine-based dispatching rules. Sublot and dispatching rule configurations are determined simultaneously by a simulated annealing-based simulation-optimization approach. Customer-oriented dispatching rules are proposed to ensure the prioritization of orders from key customers. In addition, multiple customer segments with different importance weights, their expectations and penalties on tardiness, earliness and order completion rate on due date are considered and a customer-focused objective function is formulated. In order to provide a well-adjusted structure in terms of satisfaction levels of different customer segments, weight setting functions that dynamically compute the weights in the proposed customer-oriented dispatching rules are defined. It is aimed to determine near-optimal values of the segment-based parameters of the weight setting functions. To this aim, differential evolution-based simulation-optimization approach is used. The results reveal that the proposed DSS provides more effective use of resources in satisfying customers, and can easily be implemented by manufacturing companies in practice by adopting their demand structure, customer base, customer weight settings and processing features.