Optimize Cheapest Iron Ore Blend
Budget: ₹600 – ₹1,500 INR
I have a detailed dataset covering dozens of domestic and imported iron-ore fines and lumps: their unit prices, Fe/SiO₂/Al₂O₃/P chemistry, stock-on-hand, monthly supply contracts and the logistics limits for rail, port and plant.
My goal is simple: identify the lowest-cost blend that still satisfies every operational rule we live with day to day.
The solution must respect:
• chemical envelopes for every element we report to the plant (Fe min / SiO₂, Al₂O₃, P, LOI and basicity ranges)
• the tonnage that each supplier can actually ship (supply constraints)
• the daily and monthly throughput ceilings of our yard, reclaimers and sinter plant (capacity constraints)
• logistics realities such as train rake sizes and the “domestic vs import” streams we track separately
• a hard cap on how many individual ores can appear in any single blend recipe
I am comfortable whether you build the optimiser in Excel Solver, Python (PuLP/Pyomo), or any other transparent tool—so long as I can adjust parameters later without rewriting code.
Deliverables
1. A working optimisation model that returns the minimum-cost blend for a user-entered target tonnage.
2. Clear instructions (or in-sheet notes) so I can update ore qualities, prices and limits myself.
3. A short validation run showing that the model hits every constraint and flags infeasible scenarios gracefully.
If you have prior experience with mineral blending or linear programming models, this should be a quick win; please tell me which platform you prefer to use when you reply.
My goal is simple: identify the lowest-cost blend that still satisfies every operational rule we live with day to day.
The solution must respect:
• chemical envelopes for every element we report to the plant (Fe min / SiO₂, Al₂O₃, P, LOI and basicity ranges)
• the tonnage that each supplier can actually ship (supply constraints)
• the daily and monthly throughput ceilings of our yard, reclaimers and sinter plant (capacity constraints)
• logistics realities such as train rake sizes and the “domestic vs import” streams we track separately
• a hard cap on how many individual ores can appear in any single blend recipe
I am comfortable whether you build the optimiser in Excel Solver, Python (PuLP/Pyomo), or any other transparent tool—so long as I can adjust parameters later without rewriting code.
Deliverables
1. A working optimisation model that returns the minimum-cost blend for a user-entered target tonnage.
2. Clear instructions (or in-sheet notes) so I can update ore qualities, prices and limits myself.
3. A short validation run showing that the model hits every constraint and flags infeasible scenarios gracefully.
If you have prior experience with mineral blending or linear programming models, this should be a quick win; please tell me which platform you prefer to use when you reply.