Senseible

The Firozabad glass cluster profile's fuel bill column does not add to its printed total: which row is wrong?

The rows of the Firozabad profile's annual energy bill, from The Energy and Resources Institute (TERI), add to 6,825 million rupees (Rs), not the printed 7,185, a gap of 360. The energy column does add: 309,999 against 310,000 tonnes of oil equivalent (toe). I cannot tell which bill row is wrong, so quote the energy column.

The rows

The sums

Which row is wrong

I cannot tell. The profile gives no operating days. Gas at 879,280 x 365 is 320.9 million standard cubic metres, so I used a 365 day year where a daily figure is turned into tonnes. The prices below divide each row's annual bill by its annual toe, so operating days do not enter them. Implied bill per toe, from the profile's own rows:

Why one small row does not close the gap

The 360 is bigger than the coal, pet coke, liquefied petroleum gas and kerosene rows taken one at a time. If the coal row alone were wrong, it would have to read 382, about Rs 8,650 a tonne on 44,165 tonnes, dearer than pet coke at about Rs 4,795. So one wrong coal row may not be the whole story. The gap could sit in the gas row, where it is 5.8%, or be spread across rows.

What to quote

The energy column adds up, the bill column does not. For what gas or coal cost you, use your own invoices, not the cluster table.

Implied price of coal and pet coke

A 365 day year is my assumption. At 330 days both prices rise by the same 10.6%, so the ratio does not change.

What Senseible does here

Senseible reads fuel invoices and electricity bills into monthly records, so once you upload your gas, coal and diesel bills they sit month by month, which is the figure to use in place of a cluster table.

Sources: TERI, Cluster profile: Firozabad glass industries (SAMEEEKSHA)

Last reviewed 2026-11-06. Figures are from TERI's 2015 cluster profile, which describes the cluster at that time; fuels, kiln counts and prices have moved since. Treat them as typical, and check your own records before quoting.