AIToday
Large Language ModelsTop Companies' AI MovesAI Business & IndustryTop Companies AIPublished: Nov 20, 2026, 06:31 JST

SoftBank engineer: four reasons AI can't automate LP gas delivery

Send on LINE
SoftBank engineer: four reasons AI can't automate LP gas delivery

3 Key Points

  1. What happened

    In an article by SoftBank engineer Shinya Shikishima, the company laid out four reasons LP gas delivery is hard to automate, including that it is infrastructure that cannot be stopped and that gas outages must be avoided.

  2. Why it matters

    Because supply cannot be paused and outages are unacceptable, the usual IT approach of rolling out a system and improving it in operation does not apply, so cautious verification is required.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Shinya Shikishima, an engineer at SoftBank, wrote the piece as an explanation of why AI adoption in LP gas delivery has been slow despite rapid advances in generative AI. Rather than treating the technology as a cure-all, it walks through four concrete constraints: the service cannot be stopped, gas outages must be avoided, routing combinations grow astronomically, and real-world data is messy.

A recurring theme is that efficiency cannot be pursued in isolation. Safety margins in demand forecasting mean deliveries may be scheduled earlier than strictly necessary, and extending intervals to save cost raises the risk of running out. LP gas delivery is contrasted with factories where sensors and data are relatively well organized, and data cleaning and operations matter as much as the algorithm.

SoftBank's own interest is stated at the end: it says it is advancing initiatives to support LP gas operators using communications infrastructure and digital technology. The piece closes by suggesting that time freed up from delivery and meter reading could let regional operators focus more on customer relationships, a possibility the company frames as part of its broader support for local businesses.

FAQ
Why can't LP gas delivery simply copy the usual IT approach of trying a system and improving it later?
Because LP gas delivery is social infrastructure where a supply stoppage would significantly affect customers' daily lives and business activities, the try-first approach does not necessarily apply.
What makes delivery route planning so hard?
The Vehicle Routing Problem (VRP) is cited, where the number of possible visiting orders explodes as stops increase, and comparing every combination is not realistic even for a few dozen stops.
Top Companies AIRead Original Article
Send on LINE

Also reported by ASCII.jp

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleAWS Marketplace lets AI agents buy software on a set budget