infrastructure · health
the stove is not a shelf
In Finland about half of stove fires start from things left on the stove, not food; stove guards, now required only for new stoves in senior and special-needs housing, are certified against cooking-oil fires and in tests alarmed late when a pizza box or pot holder hid the hot plate
Source: Hostikka et al., “Liesipalojen syttyminen, vaikutukset ja ehkäisy” Open ↗source: primaryLast updated: 2026-11-17
Problem statement
In Finland, cooking causes about a third of fires in homes. The country's stove-fire research project (Aalto University, the Emergency Services Academy, the safety agency Tukes and the national rescue association SPEK, 2020–2022) found that about half of these fires do not start in the food. They start in things that do not belong on the stove: kettles, coffee makers, plastic tubs, microwave covers, cutting boards, pot holders and food packaging. People use the stove as a shelf and a worktop, and a plate gets switched on by mistake or left on. The country's accident investigation board found the stove-fire situation unchanged for at least 20 years and recommended safer stoves and stove safety devices; the project's 2024 report found those recommendations had not yet been put into practice to any significant degree. The safety device on the market, the stove guard (liesivahti), is tested and classified under a European standard (EN 50615) that in its current form is "solely based on cooking oil fires." In the project's laboratory tests the guards alarmed before ignition in almost every case, but alarmed late when a pizza box or a pot holder hid the hot plate from the sensor. Guards are now required in some places. Norway has required a fixed stove guard in new electrical installations since 2010. The 2022 update of Finland's electrical installation standard (SFS 6000:2022, binding once the safety agency Tukes adds it to its standards list) requires an EN 50615 guard on new electric stoves in senior, assisted-living and special-group housing. Elsewhere it only recommends a guard; for stoves fixed in homes it requires one of several "special measures" (our translation), which, depending on the user, may also be an induction hob with controls that resist accidental switch-on, a timer or an alarm. These rules leave two gaps: they reach new installations, not the old stoves already in homes, and the guards they require are certified against the oil-pan test, not against items left on a hidden plate. The unsolved problem is preventing stove fires in the homes where they happen: homes with older cast-iron or ceramic stoves, often lived in by older people, people with memory loss or people who have been drinking. That means dealing with the stove used as a shelf, and not only with the pan of oil.
Why this matters
Between 2011 and 2020, Finland's rescue services recorded 9,422 building fires caused by cooking, a number equal to 31.8% of the fires in residential buildings (the report calls this share indicative, because some cooking fires start in other buildings). Calls for fires and fire hazards from cooking fell only slowly, from 1,161 in 2009 to 858 in 2022. In the same ten years, 73 of the 610 people who died in fires were found in the kitchen, and Nordic fire statistics attribute 51 deaths to "careless use of an electric stove" (our translation). The official numbers undercount: in the project's public survey, 73% of stove-fire incidents were never reported to the emergency number. In one insurer's claims for electric-stove fires (LähiTapiola, 2017–2020), about 68% had not been reported either. The report notes that studies have found fire-safety weaknesses piling up among people in weaker socio-economic positions. Incidents recurred more often among people whose functioning was reduced by age, memory disease, disability, or substance or mental-health problems, and they were less often able to put the fire out themselves.
What’s been tried and why it hasn’t worked
- Accident-board recommendations, mostly not carried out. The Safety Investigation Authority (OTKES) found in its stove-fire investigations that "the stove-fire situation has been the same for at least 20 years" and blamed missing or inadequate safety technology that lets human errors add up to a building fire (our translation). It recommended that stoves have at least switches that prevent accidental switch-on, and preferably induction and a stove guard. It also asked the Finnish electrotechnical standards body to push for revised stove safety standards in the EU that allow for children, older people and cooking while drunk. The 2024 project report concludes that these recommendations "have not yet been significantly implemented" and "have not yet been put into practice" (our translation). Safety technology was in use in only 11% of incidents in the public survey (n = 101). Most rescue professionals' cases (n = 35) involved an older cast-iron electric stove with no safety device or alarm.
- Stove guards certified only against oil fires. Stove guards watch the hob, sound an alarm and cut the power (EN 50615 category B). In 44 laboratory tests with two commercial guards (Safera and Innohome, both EN 50615 compliant), they alarmed before ignition in almost all cases, and for oils and similar fuels the authors found them "highly reliable" at detecting overheating; they conclude that guards combined with induction hobs "should provide an effective protection against cooktop fires." The exceptions were pizza boxes and pot holders, which blocked the sensor's view of the heating plate and delayed the alarm (in one case no alarm came). The authors did not test whether ignition was actually prevented, because the guards were modified not to cut the power. Solid kitchen items ignited with probability 0.80 ± 0.10, within 378 ± 228 s; oils and butter with 0.44 ± 0.13. Melting polyethylene bags burned at over 2,000 kW/m². The authors conclude that "further development of the stove guard standard is therefore necessary" to prevent fires from items other than cooking oil. The amendment now in draft (EN 50615:2015/prA1:2026, at enquiry in April 2026; only its public sample was read) lists its significant technical changes as updated references, a link to the Low Voltage Directive, and new marking, heating, construction and rusting requirements; non-oil test items are not among them. The part of its rationale shown in the public sample cites the Finnish project for smoke and toxic gases without ignition, as a possible new Category C. Finland's guard mandate references this standard, and in Norway both Bergen's fire service and the national civil-protection directorate (DSB) tell buyers to check that a guard is tested to it, so the oil-only test is what mandated guards are built to pass. The US standard has the same narrow aim. NIST describes UL 858 as the only one of the cooktop ignition-prevention standards it lists that addresses electric-coil cooktops, a cast-iron-pan oil or dry-pan test. In NIST's tests, small changes such as the pan's bottom finish changed the heat reaching the limit sensor and whether oil ignited.
- What the data show people actually do. Fire-service records for 2016–2020 rank what caught fire on the stove: fat 26%, plastic containers 17%, assorted items 11%, coffee makers 9%, kettles 8%, then microwave covers, food packaging, cutting boards and pot holders at about 4% each. A third of public-survey incidents (33%) began with a plate switched on by accident. In the professionals' cases, 49% involved a plate that was on by accident, often because of a pet (26%). The professionals' answers point to kitchen design: there is not enough non-flammable counter space, so the stove serves as storage and worktop, and a microwave cover or cutting board needs somewhere else to go. An estimate in the 2024 slides by three of the project's researchers, based on the fire-service shares of what ignited, splits stove fires into those an induction hob would prevent (28%), those a stove guard would prevent (50%, the food and fat fires, assuming the guard reacts in time to them), and those neither would prevent (22%); together the two would avoid an estimated 78%. The estimate also assumes that half of the other materials are ones that would ignite.
What would unlock progress
Two changes, both aimed at the stove as people really use it. First, a guard test that includes the items that actually catch fire (plastic tubs, kettles, pot holders and cardboard covering the plate), so that devices are designed and certified against hidden heating, not only an open pan of oil. Second, kitchen objects that remove the reasons a stove becomes a shelf and gets switched on by accident: controls that a hip, a sleeve or a cat cannot turn on, and a heat-safe place beside the stove for the kettle, the cutting board and the microwave cover. Because the report calls for landlords to buy safer stoves and for funding to raise safety in old buildings, and the new guard rules reach only new installations, whatever is designed has to be cheap to retrofit onto the stoves already there and has to keep working. The report notes that the upkeep of safety technology already installed also needs attention.
How we’d know it’s being solved
The national measure is the yearly number of rescue-service calls for cooking fires and fire hazards (PRONTO), falling faster than the slow decline from 1,161 (2009) to 858 (2022), together with the share where something other than food ignited first. The undercount is large, so the national measure needs a check from outside the 112 records: a repeat of the public survey and of the insurance-claim analysis. For a device or standard, the test is detection and power cut-off before ignition with the plate hidden under the item types in the fire-service list, not only with an open pan of oil. For a kitchen object, the test is fewer items found on the stove and fewer accidental switch-ons in observed kitchens over several weeks.
Where to start
Facility-free door (stove-top audit): with permission, photograph and log what sits on or within reach of the stove in shared student kitchens and in the kitchens of older volunteers reached through a senior centre. Do it at the same time of day for two weeks, and note the available counter space and the position of the controls. Compare the item list with the Finnish fire-service ranking (kettles, plastic tubs, coffee makers, microwave covers, cutting boards, pot holders). Deliver a ranked list of the items and kitchen layouts that most often turn the stove into a shelf. Needs nothing but consent and a camera.
Make-and-test door (landing space, then controls): stove guards already exist, so this door does not build one; it works on why things are left on the plate and why the plate gets switched on. The object to make is a heat-safe landing shelf or tray that fits beside an older freestanding stove where there is no non-flammable counter, sized for the items in the audit (kettle, cutting board, microwave cover, pot holders) and cheap to fit in a rented flat without tools or drilling. Build it in a campus makerspace from non-combustible material, and judge heat safety from the material's rated temperature, not by heating items on it. Check fit and reach with older adults at a senior centre and with students, using an unpowered or switched-off stove; then place it in shared student kitchens for two weeks and repeat the stove-top audit to see whether fewer items end up on the plate. Second, the controls. Protecting them is not new: the accident board recommended switches that prevent accidental switch-on, the rescue professionals listed control protection among their preventive measures, and retrofit knob covers and hob child locks are likely already on the market (no read source evaluates them). So test what exists before designing anything: on an unpowered stove, measure the time and force for an older cook to switch on deliberately and count accidental switch-ons in a simulated hip, sleeve or pet bump test. Design a new cover only for a failure the test finds. If the team wants to include people with memory loss, the deliverable is the test design, and it needs ethics review.
Lab door (access line): extending the guard test to hidden-plate scenarios means heating real items to ignition. That needs a fire laboratory with a hood or calorimeter, such as the one the Aalto team used. A team without that access should write the test protocol (item list, placement, plate types, pass criteria) from the fire-service ranking and the published test results, and hand it to a fire-safety lab or standards committee.
Relevant skills: product design, human factors, kitchen and interior design, fire-safety engineering (lab door only), observational study design.
Genes
Related problems
Stuck the same way, in other fields.
field: society & learningsixty-five thousand acres burned for dinnerRefugees spend $2 billion a year on firewood while forests are cut down around their settlements — and clean alternatives keep failing at adoptionbuilt without the people who'd live with itsource: primaryfield: making & materialsfactory data in a thousand locked drawersManufacturing data trapped in proprietary silos blocks smart manufacturing adoption by small and medium manufacturersbuilt without the people who'd live with itsource: secondaryfield: food & the living worldthe kilns changed, the workers didn'tIndia's 50,000 brick kilns must convert to cleaner technology by mandate — but 43% of converted kilns surveyed in bihar operate inefficiently because the workforce transition was never designedbuilt without the people who'd live with itsource: secondaryfield: health & medicinewrong cells, published anywayCell line misidentification persists despite decades of warningsbuilt without the people who'd live with itsource: secondarySources
- Hostikka, S, Castagnoli, E, Mikkola, R, Salonen, H, Jhatial, T, Törrö, R, Kallada Janardhan, R, Aatamila, M, Laitinen, J, Hassinen, M, Toivanen, P, Kiviranta, K, Kuurne, L, Ojala, T, Lepistö, J. "Liesipalojen syttyminen, vaikutukset ja ehkäisy." Aalto-yliopiston julkaisusarja TIEDE + TEKNOLOGIA 2/2024, Aalto-yliopisto, Helsinki 2024, ISBN 978-952-64-1803-2 (pdf), accessed 2026-11-17. Link ↗ Link 2 ↗
- Hostikka, S, Jhatial, T, & Aatamila, M. (2023). Exploring the thermal characteristics, ignitions and heat release of oils and solid items at electric cooktops. Fire Safety Journal, 140, Article 103889. (open copy, CC BY), accessed 2026-11-17. Link ↗ Link 2 ↗
- Aatamila M, Ojala T, Kuurne L. "Liesipalojen syttymissyyt, ehkäisy ja kustannukset." Pelastustoimen tutkimus- ja kehittämispäivät 5.-6.6.2024, Pelastusopisto (slides), accessed 2026-11-17. Link ↗
- Hamins A, Kim SC, Bundy M. "Investigation of Residential Cooktop Ignition Prevention Technologies." NIST Technical Note 1986, May 2018, accessed 2026-11-17. Link ↗ Link 2 ↗
- Linja-aho V. "Sähköasennustandardi SFS 6000 päivittyi." ETN, 24.08.2022, accessed 2026-11-17. Link ↗
- Bergen kommune, "Komfyrvakt" (Brann og redning), accessed 2026-11-17. Link ↗
- CENELEC CLC/TC 61, EN 50615:2015/prA1:2026 (enquiry draft, April 2026), public sample of SIST EN 50615:2015/oprA1:2026, accessed 2026-11-17. Link ↗
- Direktoratet for samfunnssikkerhet og beredskap (DSB), "Dette bør du vite om komfyrvakt" (read in Norwegian), accessed 2026-11-17. Link ↗
- CENELEC CLC/TC 61, EN 50615:2015/prA2:2026 (enquiry draft, April 2026), public sample of SIST EN 50615:2015/oprA2:2026, accessed 2026-11-17. Link ↗
- Kuurne L. "Liesipalot vakuutusyhtiölle tehtyjen vahinkoilmoitusten näkökulmasta (Tutkimusraportti)." SPEK tutkii 30, Suomen Pelastusalan Keskusjärjestö 2023, ISBN 978-951-797-728-9 (read in Finnish, identical to a saved copy; the report's own record gives the publisher's publications page), accessed 2026-11-17. Link ↗ Link 2 ↗
Revision notes
This brief gets checked against its sources and contains the following revisions. Future revisions get added here.
Worth knowing
- Several figures come from small, self-selected surveys (101 members of the public; 35 cases from rescue professionals).
- The rescue-service and insurance figures cover electric stoves only.
Full check
Sources read
- Hostikka et al., stove fires project report (Aalto University, 2024): full text (read in Finnish)
- Hostikka, Jhatial and Aatamila, ignitions at electric cooktops (Fire Safety Journal, 2023): full text
- Aatamila, Ojala and Kuurne, Emergency Services Academy slides on stove fires (2024): full text (read in Finnish)
- Kuurne, stove fires in insurance claims (SPEK, 2023): full text (read in Finnish)
- Hamins, Kim and Bundy, cooktop ignition prevention (NIST Technical Note 1986, 2018): full text
- Linja-aho, ETN article on the updated electrical installation standard (2022): full text (read in Finnish)
- Bergen kommune, stove guard page: full text (read in Norwegian)
- Draft amendment prA1:2026 to the stove-guard standard EN 50615: full text (public sample pages only; the rest is paid)
- EN 50615:2015, the stove-guard standard: could not be opened (paid; described as the papers describe it)
What changed
- Not found in the sources, replaced with what they say: that the people most at risk live in older rented homes; the page now gives the report's call for landlords to buy safer stoves and for funding to raise safety in old buildings.
- Stated more strongly than the source: that the draft amendment cites the Finnish project only for smoke and gases; the public sample breaks off partway through that passage; that NIST calls the US standard UL 858 the only one addressing electric-coil cooktops; NIST says so only among the standards it lists.
- Numbers: the 31.8% share of residential fires caused by cooking now carries the report's caution that the share is indicative, since some cooking fires start in other buildings; the 49% of rescue professionals' cases now says the plate was on by accident, which can include a plate left on, not only one switched on.
- Credit: the split of stove fires that induction or a guard would prevent (28%, 50%, 22%) is now credited to conference slides by three of the project's researchers, not to the project report.
Left open
- what the stove-guard standard and its draft amendments test [paid; only public sample pages were read]
- when Finland's updated electrical standard became binding [no read source gives the date]
Separate check of specific claims: Norway's national civil-protection directorate's advice to check that a guard is tested to the European standard was added, the page now says only the public sample of the draft amendment was read, a second draft amendment (on gas hobs) is described, and citations were added to the source list.
Earlier full check: Seven details were corrected, including when Finland's guard rule becomes binding, the other measures it allows, a translated quotation now marked as ours, the unreported insurance claims now credited to one insurer, what the Norwegian rule is said to reference, and UL 858 called a standard rather than a rule.
Separate check of specific claims: Entry points: the build starting point now makes a heat-safe landing shelf beside the stove rather than a new control cover, tests existing control covers on an unpowered stove, judges heat safety from material ratings without heating items, and adds an ethics line for people with memory loss.
Earlier full check: Stove-guard requirements in Norway and in Finnish senior housing were found and the page now starts from what they leave out, a claim about what a guard alone would prevent was toned down, the guards' good results with oil fires were added, and the long title was changed (display title unchanged).