The short version. PNGRB's Expert Committee on Optimisation of LPG and Petroleum Product Movements, chaired by Shri Arvind Kumar and assisted by Deloitte, reported in July 2026 that only 45% of India's LPG moved by pipeline in FY2023-24, with the rest split between road (about 58% by volume, more in some regions) and rail; the North-East has no LPG pipelines at all. For petroleum products the picture flips — pipelines dominate nationally, with rail at around 29%, coastal shipping at around 10% and road at only around 1% — but the committee's own pipeline-by-pipeline table shows that average conceals huge swings: from a Panipat–Delhi line running at 7.9% of capacity to a Haldia–Mourigram–Rajbandh line running at 165.3%. This piece reports what the committee's own tables say, pipeline by pipeline, and what it recommended.

45%Share of India's LPG moved by pipeline, FY2023-24 (rest: road and rail)

29% / 10% / 1%Petroleum products moved by rail / coastal / road nationally, remainder by pipeline

7.9%–165.3%Range of individual pipeline utilisation, FY26, across 56 pipelines the committee tabled

52 pipelinesIndia's total operating crude/product pipeline count, per the committee's own summary table

What the committee was asked to do

PNGRB set four terms of reference: optimise pipeline usage on an industry basis; minimise road transportation; ensure uninterrupted supply during unforeseen exigencies; and develop pipeline infrastructure on an industry basis. The committee worked with the oil marketing companies (IOCL, HPCL, BPCL, and others) to assess storage capacity, product-exchange arrangements and supply plans, covering both common-carrier pipelines (open to more than one shipper) and proprietary, single-company lines. Its brief, in its own words, was "to locate bottlenecks, identify assets that are underused, and set out how the pipeline grid should be expanded and integrated."

LPG: still mostly a road business

The committee's national tally for FY2023-24: roughly 15,665 thousand tonnes of LPG moved by road, 2,166 thousand tonnes by rail, and 13,455 thousand tonnes by pipeline — a 45% pipeline share nationally, which the report frames as "significant scope to expand pipeline connectivity." The regional picture varies sharply: the southern region ran close to an even split (49% pipeline, 48% road, 3% rail in the year the report examines) after the Mangalore–Hassan–Mysore–Bangalore pipeline came online, though the report separately notes Bangalore's BPCL bottling plant still receives roughly 40% of its LPG by road truck because it isn't connected to that pipeline. The northern region ran 34% pipeline, 51% road and 15% rail, served by the common-carrier Jamnagar–Loni line and Indian Oil's self-use Panipat–Jalandhar line. The western region ran a nearly even 43% pipeline, 50% road split (with 7% rail) in Maharashtra specifically, and the report notes Goa's bottling plants receive 100% of their LPG by road tanker from as far away as Mangalore. The eastern region was the most road-dependent the report measured, at roughly 57–58% by road, and states plainly that "the North-Eastern region [has] no pipelines," with HPCL trucking the entirety of one location's supply nearly 1,000 km from Haldia port.

The committee's fix is the Jamnagar–Loni line's own history in miniature: it reports the pipeline's capacity was raised from an original figure to 3.5 million tonnes a year, with a further augmentation to 6.5 million tonnes a year proposed specifically to reduce road dependency in Rajasthan and Uttar Pradesh and to connect it with Indian Oil's Tikri Kalan LPG plant. Three more common-carrier LPG pipelines are under construction: the Kandla–Gorakhpur line (IHB, 2,805 km, 8.25 million tonnes a year, partially commissioned as far as Jhansi), the Haldia–Panagarh line (HPCL, 215 km, 1.45 million tonnes), and the Kochi–Coimbatore–Erode–Salem line (KSPPL, 429 km, 1.53 million tonnes, partially commissioned as far as Palakkad). PNGRB has separately suo-motu bid out four further LPG pipelines: GAIL's Jhansi–Sitarganj (611 km), Shikra–Hubballi–Goa (633 km) and Cherlapally–Nagpur (556 km) lines, and a Paradeep–Raipur LPG line awarded to Dilip Buildcon (555 km).

Petroleum products: pipeline-dominant nationally, rail still essential

For petrol, diesel and other refined products, the committee describes a modal mix "dominated by pipelines followed by railways and coastal mode with road transport accounting for only a minor share (about 1% only)." Rail carries around 29% of the total, using a fleet of 220 dedicated tank-wagon rakes, and the report is candid that rail remains essential rather than a stopgap: "rail movement of petroleum products is an essential mode and cannot be entirely replaced with pipelines or coastal," because pipelines need large, consistent batch sizes and big tankage at their end terminals, while rail can serve smaller or more variable destinations. Coastal shipping carries around 10%, serving 65 terminals, and the report flags port-jetty congestion and high terminalling costs as constraints on expanding it further. Road, at roughly 1%, is reserved mainly for hilly terrain (the north-eastern states, Sikkim, Ladakh, Jammu & Kashmir, Uttarakhand, Himachal Pradesh) where rail and pipeline don't reach. Overall the network runs through 142 pipeline-fed terminals, 105 rail-fed, 65 coastal-fed and 55 road-fed, with Indian Oil operating the largest terminal network, ahead of BPCL and HPCL.

On demand, the committee cites FY2025 consumption of roughly 40 million tonnes of petrol, 91 million tonnes of diesel and 9 million tonnes of ATF, with petrol demand growing fastest (about 7.7% a year over the past decade), diesel slowest (about 2.8%) and ATF in between (about 4.6%, after a pandemic-era collapse). It projects combined petrol, diesel and ATF demand could exceed 150 million tonnes by 2030.

The pipeline-by-pipeline table: from 7.9% to 165.3%

The most useful table in the report, for this blog's purposes, is a line-by-line utilisation register for FY26 covering 38 "dedicated" (single-company) product pipelines and 18 common-carrier pipelines. It resolves several things this blog's companion pieces on ATF pipelines and pipeline utilisation had to leave as unverified or conflicting, because this piece can now cite the same register those pieces could only describe secondhand. Some of the extremes and the pipelines this blog has covered before:

Utilisation over 100% means a pipeline is moving more than its own declared nameplate capacity, which the committee does not treat as unusual: several of the highest-utilisation lines in the table are short ATF or refinery spurs originally sized for a smaller, single customer's demand that has since grown past what the pipeline was built for. The pipelines this blog previously described as feeding Lucknow, Kolkata, Kochi, Chennai, Mumbai (both a BPCL and a separate HPCL line, not one), Bengaluru, Jewar and Navi Mumbai airports all appear here by name, resolving in particular whether Lucknow has a real, currently-listed ATF pipeline (it does: a 6 km Indian Oil line, PNGRB's own register confirms) and what the Piyala–Jewar line's capacity actually is (4.58 million tonnes a year, exactly as this blog's ATF piece had found in search results but could not verify against a document).

The committee's summary table for the whole product-pipeline network, corrected from what this blog's earlier pieces could only reconstruct from search snippets: 52 pipelines in total, 20,919 km, 177.7 million tonnes a year of capacity. Of that, 38 pipelines (13,653 km, 110.8 MMTPA) are dedicated single-company lines and 14 (6,606 km, 66.9 MMTPA) are common carrier — a common-carrier share of roughly 31.6% by length, close to but not exactly the "roughly 30%" this blog had cited from a search snippet. Indian Oil runs the largest share: 24 dedicated pipelines (9,547 km, 60.2 MMTPA) plus 4 common-carrier pipelines (2,046 km, 16.6 MMTPA). HPCL runs 10 dedicated (2,128 km, 36.3 MMTPA) plus 6 common-carrier (2,481 km, 27.6 MMTPA). BPCL runs 4 dedicated (1,978 km, 14.3 MMTPA) plus 3 common-carrier (1,419 km, 17.2 MMTPA). Oil India runs a single common-carrier line (660 km, 5.5 MMTPA declared capacity — which does not match the 1.7 MMTPA the pipeline-level table above gives for that same Numaligarh–Siliguri line, a discrepancy inside the committee's own report that this piece could not resolve).

What the committee recommended

The committee's headline recommendations track its own terms of reference: connect nearby bottling plants to existing common-carrier pipelines rather than trucking to them (naming, for example, two Andhra Pradesh plants it says should be linked to the Hassan–Cherlapally LPG line); build new rail-unloading facilities to cut specific, quantified road movement (a new Madurai facility is projected to remove about 0.2 MMTPA of road movement fed from Mangalore and Yediyur); and push, as a taxation fix, for petrol, diesel and ATF to be brought under GST, or for the Central Sales Tax to be abolished as an interim step, on the reasoning that this would let refiners source products from the nearest available depot rather than a further one chosen only to avoid an inter-state tax difference — indirectly reducing infructuous road and rail movement. It also recommends common-carrier capacity be built to match actual measured demand rather than a shipper's own projection, and that the cost of new common-carrier capacity be shared proportionately among the companies that will use it.

What doesn't follow from any of this

A pipeline running at 165.3% or 150.6% of its stated capacity is not proof of dangerous over-operation; capacity figures in this table are the pipeline's original declared nameplate figure, and companies routinely debottleneck a line (adding pumping capacity, for instance) to move more than its original design figure once demand justifies it, which is a normal, planned response rather than a warning sign. Nor does a low figure like 7.9% or 2% mean a pipeline is a wasted asset: several of the lowest-utilisation lines in the table are recently built or expanded specifically to serve growing demand that has not yet caught up to the capacity installed for it, the same pattern this blog's companion piece on gas-pipeline utilisation describes for the trunk gas grid. And the unresolved 5.5-versus-1.7-million-tonne discrepancy for the Numaligarh–Siliguri line inside the committee's own report is a flag to treat that one figure carefully, not a reason to doubt the rest of a document this piece could otherwise verify by reading it directly.

Sources and caveats

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.