A strategic planning model for eco-efficient transactional service system with stochastic demand on a finite planning horizon

• Main Author: Uy, Janet Jill Yu
• Format: text
• Language: English
• Published: Animo Repository 2006
• Subjects: Strategic planning; Service industries > Management; Industrial Engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/3402; https://animorepository.dlsu.edu.ph/context/etd_masteral/article/10240/viewcontent/CDTG004082_P.pdf
• Institution: De La Salle University

The dominant role service industries play in modern society calls for the need to consider service as a source of environmental harm and as a potential instrument to reduce environmental impacts. Current researches reveal the absence of an integrative tool by which service managers can evaluate the economic and environmental aspects of the business simultaneously and not miss the effects of trade-offs. Thus, a mixed integer non-linear programming model was formulated for a single channel, multiphase transactional service system which contributes significantly to the overall level of activity of the service industry. The model was developed based on the sample path method. Entity-based modeling was also used to allow this simulation characteristic to address the over-simplification disadvantage of an analytical optimization model. The models objective was to minimize the total system cost which consists of the waiting cost, labor cost, maintenance cost, capital cost, and expansion cost. Eco-efficiency measures in the form of energy intensity and waste intensity of technological capacity were incorporated in the model to determine the volume of total energy consumed and total waste generated. These were then converted into cost and added to the objective function. The decision includes what type of technology to acquire and how much capacity, given stochastic demand on a finite planning horizon. Binary variables were also included in the technological constraints to allow the model to decide when investments must be made. The model was run in GAMS and Risk Optimizer. From the design of experiment, it was found out that energy intensity, waste intensity, maintenance cost, and capital cost are the major factors that affect the system. However, as the variability of the inter-arrival time and service requirement of customers increases, the waiting cost and waiting time also become significant. Total waiting time, and thus total waiting cost increases but the total cost decreases as the model tries to find just enough capacity to meet the demand. The response surface methodology revealed a number of conflicting factors. When energy intensity is low and capital cost is low, the total waste generated would be low but this would increase total maintenance cost and total labor cost. To have low total energy consumption, capital cost must be high but this would tend to increase total waste generation. With respect to the total system cost, energy intensity is the most important factor followed by waste intensity, thus proving the benefits of being eco-efficient. Aside from being able to consider the environmental aspect of the service business, the formulated stochastic model exceeds the capability of the conventional queuing formulas by being able to address the limitations of explosive systems and bottlenecks, and by being able to handle any distribution of inter-arrival time and customer requirement while still providing the right information in terms of expected waiting time and costs of the dynamic transactional service system.