Integrated Capacity and Inventory Optimization for Bulk Blending Plants
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current bulk product blending and packaging plants face challenges in optimizing capacity, production, and inventory planning due to hierarchical separation of these processes, leading to inefficiencies in cost reduction and demand capture, with existing methods failing to integrate equipment, workforce, and shift structure decisions effectively.
Innovation Solution
A method and tool that uses a mixed-integer non-linear mathematical optimization model to simultaneously evaluate long-term capacity, medium-term capacity, production, and inventory decisions, optimizing batch sizes, safety stock, and equipment usage to minimize operating costs while meeting key constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If hierarchical planning separates capacity planning from production and inventory planning, then capacity decisions can be made systematically, but the operational plan fails to capture opportunities to reduce cost and improve demand capture
Solution Approach 1:
The patent merges previously separate capacity planning, production planning, and inventory planning into a single integrated optimization model. This allows simultaneous consideration of equipment additions, shift structures, workforce levels, batch sizes, and safety stock decisions, enabling the system to capture cost reduction opportunities that hierarchical planning misses while maintaining systematic decision-making through mathematical optimization
2Ease of manufacture
If production batch size is increased to reduce changeover costs, then manufacturing cost decreases, but wait time for other batches increases requiring increased safety stock
Solution Approach 1:
The patent transforms the batch size decision from a fixed parameter into an optimization variable that is dynamically determined based on multiple factors including changeover costs, demand patterns, capacity constraints, and inventory holding costs. The mathematical model simultaneously optimizes batch sizes and safety stock levels, finding the optimal trade-off point that minimizes total cost rather than requiring arbitrary increases in either parameter
3Reliability
If excess inventory and spare manufacturing capacity are maintained to meet unanticipated demand spikes, then demand fulfillment reliability improves, but operating cost increases
Solution Approach 1:
The patent replaces static excess inventory and spare capacity with dynamic optimization that adjusts production schedules, batch sizes, and safety stock levels based on actual demand patterns and capacity constraints. The integrated model determines the minimum necessary safety stock required to meet service level requirements while considering the interaction with capacity decisions, eliminating the need for conservative excess inventory holdings
4Ease of manufacture
If medium-range capacity planning is done after long-term capacity decisions, then strategic resource decisions can be made first, but the operational plan cannot optimize production and inventory in view of actual capacity
Solution Approach 1:
The patent inverts the traditional hierarchical approach by using integrated simultaneous optimization that determines capacity requirements based on optimal production and inventory plans rather than imposing capacity constraints first. The mathematical model works backwards from demand and operational requirements to determine the optimal capacity configuration, enabling true operational optimization while still making strategic resource allocation decisions
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of generating capacity, production and inventory plans, over a designated planning horizon, which will optimize, with respect to a designated performance index, the operations of one or more bulk product blending and packaging plants while meeting key operating constraints. Also decision making tools and computer implemented programs for performing the method.