The internet of things is being used increasingly in the home, so Doctors Emma Dean, Mark Taylor and Denis Reilly set out here the fire safety requirements for remote controlled cookers

INTERNET OF things (IoT) devices are starting to be introduced into households, for example to control heating and lighting1. While internet enabled cooking devices for the home, such as internet enabled ovens, microwave ovens and rice cookers, are being developed by manufacturers2, they could potentially increase the risk of kitchen fires occurring, since they would be used in an unattended mode.
 
This article looks at the requirements for the safe use and design of internet enabled cooking devices by examining safe operation embedded software controls and mobile telephone alerts related to potentially unsafe operation.
 
In particular, we focus on a design for embedded control software that could be implemented in each type of internet enabled cooking device, to ensure that cooking could not be undertaken for a time 
and temperature combination that could result in a fire. 
 
Another area for our attention is the design of mobile telephone alerts to advise the user, in order to prevent unsafe operation instructions being passed to the cooking device and being carried out. 
 
We also explore the design of internet enabled cooking devices for use by vulnerable individuals – such as dementia patients – where unsafe use alerts could be sent to a relative/carer/neighbour, and for visually impaired users, where an inability to detect the correct button or pushing the wrong button on a microwave oven might lead to fire and/or accidents3.
 
Connections everywhere
 
The IoT allows people and physical devices to be connected at any time, in any place, with anything and anyone4. The combination of the internet and embedded sensors can allow everyday objects to be transformed into smart objects that can understand and react to their environment5. Andersson and Mattsson (2015)6 commented that in the everyday life of individuals, many products such as kitchen equipment, home appliances, and lighting and heating products have become internet enabled. 
 
In the UK and elsewhere, more fires and fire injuries are caused in the kitchen than anywhere else in the home7,8,9. Gao et al (2014)10 commented that kitchen fires are mostly caused by human negligence. In particular, leaving cooking unattended is the predominant cause of cooking fires7,11,12,13,14,15. Yared et al (2015)16 commented that unattended cooking is typically the main factor responsible for fires in the kitchen. In particular, older people and those who may have some form of cognitive impairment, such as Alzheimer’s disease (which can have a significant impact on both short term memory and on the level of insight), are potentially more at risk from such fires.
 
Other groups potentially more at risk of kitchen fires are drug and alcohol users, who may turn on the cooker and forget that they have done so, and people with conditions such as narcolepsy, who may fall asleep while cooking. It is important that cooking devices are used at the correct temperature and for the correct duration for the food being prepared17
 
As well as conventional oven fire incidences, Thambiraj et al (2013)18 commented on the increasing rate of inappropriate use of microwave ovens, potentially leading to fire. In particular, aluminium, paper and plastic food bags may cause a fire in a microwave oven. In addition, an empty microwave oven may catch fire if used, as if there are no foodstuffs to absorb the microwaves, the magnetron within the microwave oven ends up absorbing the microwaves and essentially self destructs.
 
Fire risks
 
There are different types of cooking devices that can now be controlled via the internet, including cookers, microwave ovens, and more specialist cooking devices such as rice cookers2. Roman et al (2013)19 commented that fault tolerance is an essential aspect of the design of IoT devices and applications. 
 
Although previous research has examined the generic safe design and operation of IoT devices, there appears to have been little research, if any, that specifically looks at the fire risks associated with internet enabled cooking devices, the requirements for their safe use and the design aspects of such devices to reduce fire risk.
 
With sufficiently flammable foodstuffs in an internet enabled cooking device (especially those with a high oil or fat content, or a high sugar content17,20), a fire could potentially result if unsafe cooking instructions were sent to the device. If such actions were carried out over an area or region, there could be the potential for numerous building fires. Safe operation of cooking devices includes:
 
  • not leaving cooking unattended15
  • not cooking if alcohol has been consumed or prescription drugs have been taken, as this may result in drowsiness or loss of concentration15
  • ensuring correct cooking temperatures are used and not exceeded – the average temperature for oven cooking is approximately 180⁰C17
  • keeping the oven clean, since built up fat and bits of food can start a fire15
  • ensuring empty microwave ovens are not used20
  • ensuring correct cooking times are used and not exceeded17
 
For an internet enabled cooker, embedded safe operation control logic could include: 
 
  • if cooking temperature > 250⁰C, then safety alert ‘Cooking temperature too high’
  • if cooking duration time > five hours, then safety alert ‘Cooking time too long’
  • if cooking temperature multiplied by cooking duration time (in hours) > 500, then safety alert ‘Cooking time too long for this temperature’
  • if temperature > 40⁰C at start time of cooking, then safety alert ‘Cooker still warm’ – the embedded safe operation control logic ‘If temperature > 40⁰C at start time of cooking’ could help to prevent attempts to override safe operation of the internet enabled cooker by repeatedly turning the cooker on and off and thus building up the temperature
In summary, a mobile telephone alert would be sent to the user of the internet enabled cooking device under the following potentially unsafe conditions:
 
  • if the cooking temperature value was outside a pre set range stored on the internet enabled cooker
  • if the cooking time duration was outside a pre set range stored on the internet enabled cooker
  • if the product of the temperature and cooking time was greater than a pre set value stored on the internet enabled cooker
  • if the cooker temperature was above a set value stored on the internet enabled cooker at the start time of the cooking instruction 
 
For an internet enabled microwave, embedded safe operation control logic could include: 
 
  • if weight of food < 100g, then safety alert ‘Insufficient food present for cooking’ (the 100g minimum weight for an item to be cooked in a microwave oven was based upon a minimum food weight of 50g [for example an egg without shell], and a 5g cooking receptacle [for example a microwave safe plastic bowl])
  • if microwave setting (in watts) multiplied by cooking duration time (in minutes) > 10,000, then safety alert ‘Excessive cooking time for this setting’. The weight of the food (and cooking receptacle) would be measured by a load sensor in the base of the microwave oven 
 
In summary, a mobile telephone alert would be sent to the user of the internet enabled microwave oven under the following potentially unsafe conditions:
 
  • if no food, or only a small amount of food is in the microwave oven
  • if the product of the microwave oven setting and cooking time is greater than a pre set value stored on the internet enabled microwave oven
 
For an internet enabled rice cooker, embedded safe operation control logic could include: 
 
  • if weight of rice and water < 100g, then safety alert ‘Insufficient rice/water present for cooking’ (the 100g minimum weight for the contents of a rice cooker was based upon a minimum weight of rice and water, 33g of rice and 67g of water, based upon the typical 2:1 cooking ratio of water to rice, and water being slightly denser than rice)
  • if cooking duration time (in minutes) > 150, then safety alert ‘Cooking time too long’
  • if rice and water weight (grams) divided by cooking duration time (in minutes) < two, then safety alert ‘Cooking time too long for this amount of rice/water’. The weight of the rice and water would be measured by a load sensor in the base of the rice cooker
 
In summary, a mobile telephone alert would be sent to the user of the internet enabled rice cooker under the following potentially unsafe conditions:
 
  • if no water and rice, or only a small amount is in the rice cooker
  • if the cooking time is greater than a pre set value stored on the internet enabled rice cooker
  • if the weight of rice and water divided by the cooking duration time is less than a pre set value stored on the internet enabled rice cooker
 
The unsafe use mobile telephone alerts could be sent to a specified relative/carer/neighbour for a vulnerable individual such as a dementia patient, and for visually impaired users the mobile telephone alert could include a text to speech function. 
 
Design safety features as mentioned above could easily be incorporated into the internet enabled cooking devices concerned and the mobile telephone applications that control them at minimal cost to the manufacturer. 
 
Fire and rescue services universally advise that cooking should not be left unattended. However, internet enabled cooking devices allow exactly that. Different fire risks can arise from the use of internet enabled cooking devices including:
 
  • cooking at too high a temperature
  • cooking for too long
  • cooking for an inappropriate temperature and time duration combination
  • using an empty microwave oven or rice cooker, or cooking with too little contents
In an attended cooking mode, such unsafe actions would hopefully be noticed. However, in an unattended cooking mode, such unsafe actions would be more likely to occur. 
 
How to proceed
 
In this article, we have examined how the inclusion of low cost embedded software operational controls, weight sensors in microwave ovens and rice cookers, and warning mobile telephone alerts could assist 
in reducing the cooking related fire risks that are mentioned above. 
 
These could prevent unsafe cooking instructions from being actioned by internet enabled cooking devices and warn the users that such actions are unsafe. In addition, such an approach might also benefit vulnerable individuals, since the warning mobile telephone alerts could be sent to other appropriate people, and the visually impaired could be assisted, with the warning mobile telephone alerts being ‘spoken’ via a text to speech function in the mobile telephone’s application settings.
 
In order to reduce the potential risk of cooking fires resulting from the unsafe use of internet enabled cooking devices, embedded software safe operation controls and mobile telephone alerts could prevent users from inputting unsafe cooking instructions, prevent the cooking device from carrying out unsafe cooking actions, and inform the user (or other relevant people, in the case of vulnerable individuals) if attempts have been made to manually enter unsafe cooking actions, for example by a vulnerable individual or by malicious individuals or software 
 
Dr Emma Dean is associate partner – research and evaluation at Greater Manchester Fire and Rescue Service; Dr Denis Reilly and Dr Mark Taylor are lecturers at Liverpool John Moores University. For more information, view page 5
 
References
 
1. Hive (2017) Hive – wireless product control, https://www.hivehome.com/
2. Riekki, J, Sanchez, I, Pyykkonen, M (2012), ‘NFC-based user interfaces’, in proceedings of IEEE Near Field Communication (NFC) International Workshop, 13th March 2012, Mannerheiminaukio, Helsinki, Finland, pp 3-9.
3. Riazi, A, Ying Boon, M, Bridge, C, Dain, S (2012), ‘Home modification guidelines as recommended by visually impaired people’, Journal of Assistive Technologies, 6, 4, 270-284.
4. Nolin, J, Olson, N (2016), ‘The Internet of Things and convenience’, Internet Research, 26, 2, pp 360-376.
5. Kortuem, G, Kawsar, F, Fitton, D, Sundramoorthy, V (2010), ‘Smart Objects as Building Blocks for the Internet of Things’, IEEE Internet Computing, 14, 1, pp 44-51.
6. Andersson, P, Mattsson, L (2015), ‘Service innovations enabled by the “internet of things” ’, IMP Journal, 9, 1, pp 85-106.
7. LFB (2017) London Fire Brigade, Cooking, http://www.london–fire.gov.uk/SafetyInTheKitchen.asp
8. NFPA (2017) US National Fire Protection Association, Public Education, ‘Cooking’, http://www.nfpa.org/public–education/by–topic/top–causes–of–fire/cooking
10. Gao, Y, Liu, Q, Chow, W, Wu, M (2014), ‘Analytical and experimental study on multiple fire sources in a kitchen’, Fire Safety Journal, 63, pp 101-112.
11. Lushaka, B, Zalok, E (2014), ‘Development of a Sensing Device to Reduce the Risk from Kitchen Fires’, Fire Technology, 50, 3, pp 791-803.
12. Xiong, L, Bruck, D, Ball, M (2016), ‘Preventing accidental residential fires: the role of human involvement in non-injury house fires’, Fire and Materials, 41, pp 3-16.
13. DSFRS (2017) Devon and Somerset Fire and Rescue Service, Safety in the kitchen, https://www.dsfire.gov.uk/YourSafety/SafetyInTheHome/SafetyInTheKitchen/Index.cfm?siteCategoryId=4&T1ID=35&T2ID=46
14. SFRS (2017) Staffordshire Fire and Rescue Service, Unattended cooking causes kitchen fire, http://www.staffordshirefire.gov.uk/1605.asp
15. ESFRS (2017) East Sussex Fire and Rescue Service, Fire safety at home: cooking, http://www.esfrs.org/your–safety/fire–safety–at–home/cooking/
16. Yared, R, Abdulrazak, B, Tessier, T, Mabilleau, P (2015), ‘Cooking risk analysis to enhance safety of elderly people in a smart kitchen’, in proceedings of the 8th ACM International Conference on Pervasive Technologies Related to Assistive Environments, Corfu, Greece, July 1-3, 2015, DOI: http://dx.doi.org/10.1145/2769493.2769516
17. Dinaburg, J, and Gottuk, D (2016), ‘Smoke alarm nuisance source characterization: review and recommendations’, Fire Technology, 52, 5, 1, pp 197-233.
18. Thambiraj, D, Chounthirath, T, Smith, G (2013), ‘Microwave oven-related injuries treated in hospital EDs in the United States, 1990 to 2010’, The American Journal of Emergency Medicine, 31, 6, pp 958-963.
19. Roman, R, Zhou, J, Lopez, J (2013), ‘On the features and challenges of security and privacy in distributed internet of things’, Computer Networks, 57, 2, pp 266-2,279.
20. Swain, M, Russell, S, Clarke, R, Swain, M (2004), ‘The development of food simulants for microwave oven testing’, International Journal of Food Science & Technology, 39, 6, pp 623-630.
 
Also in this issue