We focus on planning transportation operations within a blood sample supply chain, which comprises clinics and a laboratory. Specifically, the main goal of this study is to obtain the optimal number of vehicles to be deployed and the scheduling of the pickup process. First, we formulate a mixed-integer programming (MIP) problem. Next, we develop a heuristic scheme composed of two heuristic algorithms and numerical search, and a new genetic algorithm. In an extensive numerical study, based on the data from a real-life blood sample collection process, we illustrate the potential of the new heuristic scheme. highly diverse and geographically dispersed (Yi, 2003), whereas the availability of testing facilities is likely to be more limited. In addition, a high percentage of samples are collected in the morning, from customers who have fasted overnight; as a result, the delivery burden is especially high during this time of day. An additional challenge is the cycle of measuring time from taking the blood samples following the doctor's order until the results are transmitted back to the doctor (Grasas et al., 2014).The effectiveness of a blood sample collection process, in terms of operational costs, customer satisfaction, and resource utilization, depends heavily on the manner in which the corresponding healthcare system regulates and operates its blood sample supply chain. Herein, we focus on improving the effectiveness of a blood sample supply chain comprising clinics, where blood samples are collected and processed, and a centralized testing laboratory, where blood samples are analyzed (Yücel et al., 2013;Grasas et al., 2014). Specifically, we seek to optimize the process by which blood samples are transported from clinics to the laboratory, bearing in mind that, as noted above, blood samples must be transported and analyzed within a strict time constraint.We aim to determine the number of vehicles that should be used for sample delivery and to identify the optimal schedule for the pickup process (Revere, 2004;Grasas et al., 2014). For this purpose, we formulate a multiobjective mixed-integer programming (MIP) problem that minimizes the vehicle fleet size, total transportation time, and the expected quantity of collected samples delivered "too late" to the testing laboratory. In order to evaluate the effectiveness of the chain, we introduce two performance measures: operational efficiency and quality of service. Operational efficiency refers to operational costs, for example, the size of the vehicle fleet involved in the pickup and transportation process. Quality of service refers to the quantity of samples delivered to the testing laboratory within a certain time span after collection. We develop an efficient heuristic scheme to approximate a solution for this problem, as it is not feasible to obtain exact optimal solutions for MIP problemseven for small-sized instances-in a reasonable period of time. A comparison analysis between the MIP solutions and the outputs of the heuristic shows that the heuristic closely a...