Supply Chain Network Design Tool for Commodity Gas Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current supply chain management systems for cylinder gas distribution networks face challenges in determining optimal locations for filling plants and hubs, as well as efficient transportation flows, which affects operational efficiency and agility due to the complexity of integrating inventory, transportation, and investment decisions.
Innovation Solution
A computer-implemented method and network design tool that uses a two-step solver approach, combining an equivalent cylinder solver and a primary stock solver, to determine the configuration of distribution hubs and filling plants, and specify transportation flows, integrating strategic and operational decisions, and employing heuristic algorithms for large-scale network optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a two-step solver approach is used to determine optimal locations for filling plants and hubs, then the operational efficiency and agility of the supply chain is improved, but the device complexity and computational requirements increase
Solution Approach 1:
The supply chain network design problem is segmented into two distinct solving steps: first, determining optimal hub locations and secondary transportation flows; second, determining optimal filling plant locations and primary transportation flows. This segmentation allows each step to focus on specific aspects of the problem, improving computational efficiency while maintaining overall optimization quality.
Solution Approach 2:
The first step of the two-step solver performs preliminary actions by determining hub locations and secondary transportation flows before proceeding to the second step. This preliminary solution provides a foundation for the subsequent optimization of filling plant locations and primary transportation, enabling iterative improvement of the overall network configuration.
2Ease of operation
If strategic decisions for filling plant and hub locations are made separately from operational decisions, then the ease of operation is improved, but the manufacturing precision and optimization quality deteriorate
Solution Approach 1:
The two-step solver merges strategic location decisions with operational transportation optimization by integrating both decision types into a unified optimization framework. The solver simultaneously considers facility locations, inventory levels, transportation flows, and service level requirements to determine the optimal network configuration, eliminating the trade-off between decision simplicity and optimization quality.
3Adaptability or versatility
If the network structure is simplified to improve agility, then the ease of operation is improved, but the loss of information about complex transportation flows increases
Solution Approach 1:
The two-step solver employs dynamic optimization by allowing the network configuration to adapt to varying conditions such as demand fluctuations, transportation costs, and service level requirements. The solver iteratively adjusts hub and filling plant locations, inventory levels, and transportation flows to optimize performance under different scenarios, maintaining agility while preserving critical flow information.
Data Source
AI summary
Techniques are disclosed for designing a supply chain distribution network. A network design tool may be configured to facilitates strategic decision making for a producer/distributor of a commodity products (e.g., molecular gas mixtures stored in cylinders) by incorporating transportation costs (i.e., operational decisions) into the strategic decision making process. Further, the network design tool may also integrate investment decision costs, which are usually considered ‘tactical decisions’, into the strategic decision process (e.g., which filling tools are assigned to which plant/hub location). The network design tool implements different algorithm approaches allowing a user to obtain optimal, near-optimal, and approximate solutions.