Conference Agenda

In this study, we focus on a capacitated lot sizing problem (CLSP) with stochastic times and limited inventory. The classical CLSP aims to minimize the total cost including the costs of production, setup and inventory, and to meet the demands using a single machine. This machine can produce several different items in each time period, where its capacity is consumed both by production and by setup operations. In real-life applications, uncertainties, especially the ones in production and setup times, are real concerns for companies. To obtain efficient production plans to be performed in real-life environment, we consider stochastic setup and production times following a given probability distribution. In this setting, overtime costs are incurred in case the machine is used beyond its production capacity. The proposed problem also addresses the limited storage space of the warehouses. More specifically, the level of the inventory at the end of each time period is assumed to be bounded to better reflect the real-life applications in which warehouses with limited storage areas are mostly used to store several different items. The problem described above is formulated as a stochastic programming with recourse (overtime) decisions. Two approximate solution approaches are developed. The first approach is based on the tabu search method, whereas the second approach is based on solving the stochastic programming with a set of sample scenarios. The computational experiments are conducted on well-known problem instances and extensive analyses are provided.

Uncertainties in demand and supply dynamics pose significant challenges to inventory planning, often leading to inefficiencies and increased costs. Traditional approaches tend to fall short in effectively addressing these challenges by accommodating fluctuations in demand. To address this issue, we propose and evaluate a novel approach utilizing a stochastic mathematical optimization model.

The key outcome is the development and implementation of a mathematical optimization model which is specifically designed to address the complexities of inventory planning by incorporating uncertainties in demand and other relevant factors such as lead times, storage constraints and purchasing costs. Furthermore, an extensive evaluation study is conducted to assess the performance of the proposed model. Various key performance indicators (KPIs) like purchasing costs, stock level and service level are analyzed to compare the effectiveness of the derived purchasing strategy with alternative approaches. Through this evaluation, insights into the model's ability to mitigate inventory-related challenges and improve operational efficiency are gained. Additionally, the study investigates the impact of parameter values, providing valuable insights into optimizing decision-making processes.

In conclusion, this study presents a comprehensive solution to the challenges of inventory planning through the development and implementation of a stochastic optimization model. By addressing uncertainties in demand and conducting a thorough evaluation of different strategies and parameters, the study contributes to improving inventory management practices and enhancing supply chain performance and resilience.

This presentation presents a method for optimally allocating a limited budget to mitigate project risks. After identifying the project risks, the project manager must prioritize which risks to address. The challenge lies in the fact that the occurrence of these risks, their impacts on the project's cost, schedule, and quality/content, and even the outcomes of the risk response plan can all be random variables. The proposed method begins with a Monte Carlo simulation to model the risk impacts and the residual effects after implementing the risk responses. The simulation results serve as inputs to mathematical programming methods, which determine the optimal budget allocation among the risks, considering various objective functions.