Small Section Rolling Mill Financial Model

Description

A small section rolling mill converts steel billets into long products – rebar, angles, channels, flats, rounds – through a hot‑rolling line with a reheat furnace, roughing/intermediate/finishing stands and a cooling bed. The model captures the multi‑product reality: separate rolling campaigns for each profile, product‑specific rolling speeds and the downtime required for roll changes and guide adjustments between campaigns. It treats billet as the dominant cost driver, along with the yield losses that eat into every tonne charged.

Costs are modelled in detail: natural gas consumption in the reheating furnace as a function of throughput and product entry temperature, electricity drawn by main and auxiliary drives varying with section size and rolling pace, roll wear and regrinding cycles, labour across shift patterns, and consumables. Yield factors are split into scale loss (typically 1–2 % of charged weight) and crop‑end loss (2–4 %), both dependent on the product rolled; a built‑in by‑product revenue stream captures resale of mill scale. Billet pricing can be tied to benchmark scrap indices plus alloy surcharges, reflecting real‑world contract structures.

On the operational side, the model explicitly respects the cooling‑bed length as a throughput bottleneck – it automatically reduces the effective production rate when rolling light sections that need longer cooling time, preventing the unrealistic capacity assumptions common in simpler spreadsheets. Quality‑driven process steps such as Quenching and Self‑Tempering (QST) for rebar can be switched on, adjusting energy use and consumable costs. Maintenance downtimes, shift calendars and intermediate billet/finished‑goods inventory with holding costs round out a model that mirrors the pace and constraints of a real mill.

Modeling specifics

• Campaign‑based production scheduling: product‑specific rolling rates (tonnes/hour) and changeover downtimes between profiles; the model automatically sequences campaigns and accumulates lost production time.
• Differentiated material yield loss: scale and crop‑end percentages set by product, so margins on thin rounds are not artificially inflated by a single average yield factor.
• Energy consumption split by source: electricity (kWh per tonne) varies with product size and rolling speed; natural gas (GJ per tonne) for the furnace depends on billet entry temperature and campaign throughput.
• Cooling‑bed capacity constraint: the effective line speed is capped when rolling light sections, reflecting real‑world heat‑transfer limits and preventing over‑estimation of annual output by 15–25 %.
• Roll and consumable tracking: consumption of rolls, guides and shear blades is linked to pass usage; regrinding costs and roll‑scrap credits are computed from campaign tonnage and roll‑life curves.
• Dynamic billet and finished‑goods inventory: stock build‑up during campaign pre‑rolling, safety‑stock logic and carrying cost, avoiding the assumption that everything is sold the moment it is rolled.

What's included in the base version

• Revenue module with product mix segmentation and multi‑grade pricing
• Variable OPEX: billets (with alloy surcharge logic), energy (gas & electricity), rolls, consumables
• Fixed OPEX: labour by shift, maintenance, overheads
• CAPEX schedule for mechanical, electrical, furnace and infrastructure
• Depreciation by asset group, fixed‑ and variable‑with‑output
• Debt financing: drawdown, repayment sculpting, interest, fees
• Profit & Loss, Cash Flow (direct), Balance Sheet
• Key metrics: IRRe, NPVe, payback period, EBITDA margin, debt service coverage
• Sensitivity tables (tornado) on 10 input drivers

Common modeling mistakes

• Ignoring campaign changeover time and roll‑pass setup — effective annual throughput is overstated by 10–20 %, compressing the payback period by 1–2 years.
• Using a single yield factor for all products — revenue is overestimated by 2–5 % on lightweight sections where crop‑end losses are proportionally larger.
• Neglecting the cooling‑bed bottleneck — the model suggests a capacity that is 15–25 % above what is physically attainable for small sections, severely distorting the capex‑return trade‑off.
• Assuming constant electricity consumption per tonne across the product mix — energy cost for small rounds is underestimated, lifting EBITDA by 3–6 % and hiding true margin compression on those products.
• Omitting billet price escalation driven by scrap and alloy markets — the raw‑material cost spike in up‑cycles is missed, inflating equity IRR by 2–4 percentage points over the project life.