Optimal network design and inventory planning for clinical trial supply chains

• Main Author: Zhao, Hui
• Other Authors: Wu Kan
• Format: Theses and Dissertations
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Industrial engineering::Operations research; DRNTU::Engineering::Industrial engineering::Supply chain
• Online Access: https://hdl.handle.net/10356/82988; http://hdl.handle.net/10220/47563
• Institution: Nanyang Technological University

The clinical trial plays a key role in the competitiveness of a pharmaceutical company. It is a critical step in new drug development and characterized by long duration, high cost, and high potential return. Due to the increasing complexity of new drug development, the profit is decreasing and the pharmaceutical companies strive to reduce the cost. In this thesis, we strive to improve the efficiency in three problems of clinical trial supply chains, including the inventory allocation, network design, and production planning problems. For each of these problems, we make a discussion of the relevant literature, modeling approaches, and algorithms. Due to the patient recruitment uncertainty in clinical trial supply chains, these problems are formulated as stochastic programming models. The inventory allocation problem considers the efficient coordination of clinical drugs in clinical trial supply chains. Due to the specific characteristics of clinical trial supply chains, the inventory policy is analytically intractable. In this study, we develop an inventory allocation model to determine the distribution plan and inventory levels of clinical drugs at each location of the supply chain in each period. The total cost of transportation, holding, and stock-out is minimized in the model. Unlike classical supply chains, the demand for clinical drugs is hard to predict and stock-outs may occur due to the patient recruitment uncertainty. When the number of recruited patients at a clinic is lower than expected, the clinical drugs at the clinic can be shipped back to the distribution center for later redistribution or even shipped directly to other clinics. To solve the model efficiently, we apply Benders decomposition as the solution approach and propose two acceleration techniques. Numerical results show that our model gives a lower cost than the (s, S) policy in most cases. In addition, efficient clinical trial supply chain configurations are identified in the network design problem. Since the clinical trial duration is included in the 20 years of patent life, its reduction may increase the drug’s commercialization time under patent protection and the total sales revenue for the specific drug significantly. However, when the duration is reduced by increasing the number of clinical sites in the supply chain, the total cost is increased simultaneously. Hence, the duration and cost become a tradeoff and Pareto-optimal supply chain configurations are used to improve the clinical trial efficiency. In this study, we propose a multi-objective clinical site selection model which considers the tradeoff between time and cost of clinical trials. An efficiency curve representing the Pareto-optimal tradeoff is provided for decision makers to design the supply chain. To identify all Pareto-optimal supply chain configurations efficiently, we propose some propositions and develop an algorithm accordingly. Some optimality cuts are also derived to improve the solving efficiency of this problem. The overproduction reduction of clinical drugs is considered in the production planning problem. Overproduction is commonly seen in clinical trial supply chains and pharmaceutical companies strive to reduce it due to the significant drug production cost nowadays. When the overproduction is reduced in clinical trial supply chains, stock-outs may occur when the allocated inventory cannot meet the stochastic demand at some clinical sites. As a result, the clinical trial duration may increase. To reduce the overproduction without the duration increase in clinical trials, the inventory in the clinical trial supply chain should be coordinated efficiently. In this study, we propose a multi-objective multi-stage stochastic model to optimize the production quantity, where the clinical trial duration and the total production and operational costs are minimized. The progressive hedging algorithm is applied as the solution approach. In the numerical experiments, we study this algorithm’s performance and find the optimal production quantity of clinical drugs.