Human Centered Approach for Reducing
Household Food Waste by Tracking Fridge
Inventory and the Use of Mobile Application
Mänskligt centrerat tillvägagångssätt för att minska hushållens matavfall
genom att spåra kylinventariet och användningen av mobilapplikation
Marko Radenkovic
Marcel Laska
Faculty of Technology and Society Computer Science
Bachelor of Science in Computer Science 180 hp VT 2020
Supervisor: Reza Malekian Examiner: Mia Persson Final Seminar 4/6 2020
Abstract
Increasing amounts of food waste is becoming a problem in developed countries. This research project is about how to reduce food waste by tracking fridge inventory and the best-before date of fridge stored food by a smartphone application. Food waste occurs in several different ways. It can be found in the household, retail stores and in the catering and events sector. Food waste can also be classified as avoidable waste, possibly avoidable waste and unavoidable waste. This project focuses on the household sector and the use of a mobile application to track fridge inventory and best-before date. The purpose of the application is to try to reduce household food waste. There is previous research about reducing household food waste using mobile applications and they all take different approaches. This study focuses on building upon these previous approaches together with data gathered from our own questionnaire with Sweden as primary focus. The combined data results in our own mobile application solution that has been tested by users in Swedish households. The study’s research question is addressed by the use of the application during a period of one-week observation. Our data collection consists of the participants being interviewed at the end of the observation period. The interviews gathered information related to whether the users successfully reduced food wastage in their households.
Contents
1. Introduction 4
1.1 Scope and Goals 5
1.2 Limitations 5
2. Background 6
2.1 Household Food Waste 6
2.1.1 Avoidable Waste 7
2.1.2 Possibly Avoidable Waste 8
2.1.3 Unavoidable Waste 8
2.2 Change of Behaviour 8
3. Previous Research 9
3.1 Fridge Inventory 9
3.2 Best-Before Date 9
3.3 Grocery Shopping Tracking 10
4. Method 11 4.1 Mixed Method 11 4.2 Study Content 11 4.2.1 User Usability 11 4.2.2 Questionnaire 12 4.2.3 Application Development 13
5. Questionnaire Result and Analysis 14
5.1 Respondents 14
5.1.1 Age and Gender 14
5.1.2 Smartphone Usage 15
5.1.3 Household Situation 15
5.2 Household Food Waste 15
5.2.1 Food Waste 15
5.2.2 Food Waste Frequency 16
5.2.3 Reasons for Food Waste 16
5.2.4 Reasons Why “Best-before-date has expired” 17
5.3 Fridge Inventory 18
5.3.1 Fridge Inventory Tracking 18
5.3.2 Grocery Shopping Tracking 18
5.3.3 Grocery Shopping Problems 19
5.4 Willingness to use Mobile Application 20
5.4.1 Use Mobile Application to Track Fridge Inventory 20 5.4.2 Willingness to Take Photo of Receipt After Grocery Shopping 20
5.4.3 Household’s Grocery List in the Application 20
5.5 Conclusion of Analysis 21
6. Mobile Application Solution 22
6.1 Functionality 22
6.2 Scanning of Receipts 22
6.2.1 OCR 23
6.2.2 Shelf Life of Refrigerated Foods 24
6.3 Best-Before-Date Notifications 27
6.3.1 Status Bar and Notification Drawer Notification 27
6.3.2 Heads Up Notification 28
6.3.3 Locked Screen Notification 29
6.4 Joint Grocery Shopping List 29
6.4.1 Firebase Realtime Database 30
6.5 General Data Protection Regulation 31
7. Results 32
7.1 Interview 32
7.2 Household Situation, Usage and User Experience 32
7.2.1 Household Situation 32
7.2.2 Usage of Application 33
7.2.3 User Experience 34
7.3 Impact on Fridge Inventory and Grocery Shopping Tracking 35
7.3.1 Impact on Fridge Inventory Tracking 35
7.3.2 Impact on Grocery Shopping Tracking 36
7.4 Change of Behaviour 37
7.4.1 Noticable Change of Behaviour 37
7.4.2 Reasons Behind Change of Behaviour 38
8. Discussion 39 9. Conclusion 43 9.1 Future Work 43 10. References 45 11. Appendices 48 A. Questionnaire 48 B. Interviews 50
1. Introduction
Approximately one third of all globally produced food is wasted or thrown away every year. It occurs at all stages of the Food Supply Chains (FSC) and the scale of it depends on factors such as wealth of the certain area, closeness and access to agriculture, culture and most importantly, supply and demand [1]. In Sweden, the National Food Agency collaborated with both the Environmental Protection Agency and the Board of Agriculture and developed a plan called “More To Do More” . The goal is to reduce waste of food in all of the supply chains through clear arrangements and actions in all parts, from the farm until it reaches the customer. The plan is supposed to help Sweden achieve their part of the global-sustainable-development goal of Agenda 2030, which is that by 2030 we should have halved the global food waste per capita at the retail and consumer levels. However, there are some problems. Swedes waste, on average, 97kg of eatable food and drinks per person per year whereat food waste in households accounts for approximately 70% of the total food wasted, [2]. Therefore it is clear that the households are the part that can be affected and addressed the most. Furthermore, the problem goes beyond only being about food waste as the environment is also affected. To produce the food that is being wasted causes 2 billion tonnes of carbon dioxide to be emitted. That is almost 3% of Sweden’s total carbon dioxide emission [3].
The Swedish National Food Agency (NFA), has reviewed the food waste problem by studying consumption patterns, behaviour and attitudes [4]. Their conclusions resulted in that the high pace of society with ever-increasing expectations of material standards means that households today often suffer from a lack of time to become more involved in cooking than necessary. This means an increased risk of wastage by neglecting the planning of purchases and over-purchasing which leads to the fact that food can become too old, best before the day is misinterpreted and that all parts of the raw materials or residues are not utilized [4]. It is clear that in order to reduce the waste of food we have to look at the highest level of the food waste hierarchy, namely the prevention of food waste.
In addition, more than 90% of Sweden’s population has access to and uses a smartphone on a daily basis, a number that keeps growing every year [5]. Therefore, by creating an application to help people reduce the amount of food being wasted we will be able to reach the highest level of the food waste hierarchy but, in theory, also reach almost the entire population of Sweden.
This leads to the research questions of this paper:
● How to track grocery inventory in the fridge and the best-before-date? ● How to avoid food waste by tracking what exists in the fridge and when it
expires?
1.1 Scope and Goals
The main task of this paper is to evaluate how to approach the waste of food within households with the help of an mobile application whereas the main challenge lies within to change the approach of Swedes concerning planning of grocery purchases and over-purchasing.
Due to the limited scope of the thesis the implementation will intentionally focus on food with an “best-before date” that must be stored in the refrigerator. Furthermore, other aspects of the food waste problem, such as the health and economic aspects are not included. However, both these aspects are possible subjects for further studies since they might be in relation to how they affect the food waste problem in households. The ultimate main goal is to reduce the waste of fridge-stored food within households. In order to achieve the main goal following sub-goals must be achieved:
● Raise concerns about food waste within households.
1.2 Limitations
The following limitations were made:
● The application was only developed for the Android mobile operating system ● Due to the short time frame the study focused only on fridge-stored food
with a best-before-date
● Due to the short time-frame the users tested the application for only one week
2. Background
It is estimated that more than 820 million people are hungry globally, a number that is up from 811 million the previous year and totally it is the third increase in a row, making the reduction of food waste even more important [6]. The topics and objects introduced in this section will cover two of the United Nations Sustainable Development Goals, namely goal 2 - “Zero Hunger” and goal 12 - “Responsible Consumption and Production” [7].
Goal 2, “Zero Hunger” is stating that a profound change of both the food and agriculture system is needed in order to nourish the 820 million hungry people and also have in mind the additional 2 billion increase of the world's population that is expected by 2050. The ultimate goal is to end hunger, achieve food security and have sustainable agriculture [7].
Goal 12, “Responsible Consumption and Production” is about promoting resource and energy efficiency and sustainable infrastructure. About one third of all globally produced food ends up either rotting in the bins of consumers and retailers or is spilled out due to poor harvesting practices or transportation. The targets of this goal is to halve per capita global food waste at the retail and consumer levels by 2030 and substantially reduce food waste generation through prevention and reduction [7].
The ultimate goal is that this paper can have an impact on both these goals, for their logos see Figure 1.
Figure 1: Logos for UN’s sustainable development goal 2, “Zero Hunger” and goal 12. “Responsible Consumption and Production” [7]
The following sections will describe food waste in Sweden, both in general and in households and the three categories of food waste.
2.1 Household Food Waste
In Sweden it is the households that are responsible for the majority of the food being wasted, almost a total of 917,000 tonnes which corresponds to 70% of the total yearly waste. Out of the total waste, 384,000 tonnes were not sorted and ended in the residual waste, most commonly known as the garbage bag under the sink. However, 217,000 tonnes were sorted for biological treatment such as
digestion and composting including home composting and smaller quantities via disposers. The remaining, approximately 224,000 tonnes of food and drink are poured into the sewage [8].
Almost 28%, corresponding to about 18kg per person per year, of households’ sorted food waste is unnecessary. The same applies for the approximately 2% per person and year of food that ends up in the compost [9]. For the distribution of food groups in residual waste see Figure 2.
Figure 2: Distribution of household food waste in food groups by residual waste [9].
Another part of the household's food waste is that which is being poured into the drain. It is estimated that Swedes, per person, throw away 26kg food that ends up in the drain, where all of it counts as avoidable waste [9]. Food wasted in the drain constitutes a quarter of the total household food waste. Almost 40% of the quantities poured consisted of coffee and tea, 25 percent of dairy products and 10 percent of other beverages such as juice, carbonated or alcoholic beverages. Then came solid food waste such as rice, pasta and cereals [10].
Food waste is categorized into three different categories, avoidable waste, possibly avoidable waste and unavoidable waste.
2.1.1 Avoidable Waste
Avoidable waste is the food and drinks that we were able to eat at some point but chose not to. Example of avoidable waste is a slice of bread, an apple, meatballs or orange juice. Avoidable waste is the food waste category that can be affected and reduced the most [11].
2.1.2 Possibly Avoidable Waste
Possibly avoidable waste is the food and drinks that some people eat and some people do not. This depends on the preferences, taste, allergies and intolerances of the consumer but also if the food can be eaten prepared in one way but not in another. Examples of possibly avoidable waste is bread crusts, potato skin or pulp in orange juice [11].
2.1.3 Unavoidable Waste
Unavoidable waste is the food and drinks arising from food preparation that is and has not been edible for us humans under normal circumstances. Example of unavoidable waste are meat bones, shell of a coconut or the skin of oranges [11].
2.2 Change of Behaviour
Habits and emotions of food wasting consumers need to be changed in order to reduce or prevent waste of food. However, it needs to be done carefully, as people aware and guilty of food waste let their feelings have a negative affect on their habits meaning that they tend to waste even more food compared to people who are not aware of the effects of their actions [12]. This leads to the fact that the application cannot initiate negative feelings regarding throwing away food and waste in general but instead focuses on giving both positive feedback as well as creating a positive image about the change of behaviour of the consumer.
3. Previous Research
There are a number of mobile applications available on the market with the purpose of reducing food waste in various settings. This thesis focuses on reducing household food waste by tracking the grocery inventory in the fridge and when it expires. In addition, focus has to be on tracking grocery shopping. If grocery shopping is tracked it can help prevent over-purchasing and thus lead to less food being wasted because of the best-before date.
3.1 Fridge Inventory
There exists several prototype mobile applications developed as a result of prior research. The prototypes have their own limitations but what they all have in common is that they want to promote behaviour change in order to reduce household food waste. The way the application keeps track of food and waste will determine how easy the application is to use. Several options have been explored for managing fridge inventory but they differ in the difficulty level to implement because of various technical aspects [13].
The optimal solution for managing fridge inventory is to retrieve purchases directly from any given grocery store. The user would afterwards log in and track the specific groceries bought. There is an alternative where the user would have to invest in buying a smart fridge in order to keep track of the groceries. This would not make the application very accessible because of the cost and scare off most potential users [13]. The aforementioned options are both difficult to realistically implement. A simpler solution to implement would be to scan barcodes from groceries. This gives the user full control of what to scan, and what not to. However, this option introduces a larger workload for the user and it is also not possible to implement if the grocery stores do not grant the necessary access to their databases [14]. Ultimately, scanning the receipt after each grocery purchase could be possible. But this option comes with the technical problem of maybe not successfully retrieving all the necessary information [13].
3.2 Best-Before Date
Prior research has proven that there is no good way with the user in mind to retrieve the best-before date of groceries. There exists a manual data entry barrier that is hard to overcome. The food infrastructure in many countries does not allow information about groceries to be easily codified at the point of sale [14]. In some countries, such as the U.S., certain information can be retrieved through scanning grocery barcodes. When the barcode of a grocery is scanned, the data retrieved can be linked to corresponding databases which include information about the grocery such as the food’s name and best-before date [14]. This is not possible today in Sweden due to the limitations of the food infrastructure. This makes it challenging and unnecessarily user-intensive to retrieve the best-before date. A user-intensive process to add best-before dates may have a negative consequence such as reduced engagement to the mobile application. This will in turn affect the goal of reducing household food waste.
3.3 Grocery Shopping Tracking
Food wastage can be greatly reduced through planning. If the consumer plans the shopping well it can help reduce throwing a chunkful of food products because of the best-before date. Good planning can also give the consumer a better shopping experience since they would not have to go back to the store in case they forgot to buy something. There are several applications available that focus on tracking grocery shopping. Fridge Pal is an example of an application targeting individual consumers [15]. It provides a range of features aimed at helping the users with managing their groceries. Some of Fridge Pal’s defining features includes creating and managing shopping lists which can later also be used to add food items to the application’s inventory [15].
4. Method
Following section will describe the method used to conduct the research. Further, a motivation for the use of chosen method is presented along with a discussion of how the method was implemented.
4.1 Mixed Method
In order to achieve a strong understanding of the problem an explanatory mixed method [16] was used during this study. Mixed methods are the collections of both quantitative, namely close-ended, and qualitative, open-ended, data. It uses established methods for the collection, analysis and interpretation of both types of data meaning that both are integrated in the analysis [16].
The main reasons why the mixed method approach is a good research methodology are [17]:
● Minimizes limitations of both qualitative and quantitative approaches ● Explains quantitative results with the qualitative analysis and collection of
data
● Complements the questionnaire by integrating individuals perspectives
The aim of the study and mentioned above reasons align making the mixed methods approach suitable to use in the study. Figure 3 below describes the mixed method.
Figure 3: The mixed method approach used in this study.
4.2 Study Content
The different parts involved in this study are presented, discussed and motivated here.
4.2.1 User Usability
In order to get the Swedish population to use the app we have to understand their point of view, what their main problem regarding food waste is, the functionality of the application and how user friendly it should be. These mentioned topics are all part of a broader concept called usability. Good usability of a product or service is crucial in order to achieve success, in this case to reach out to the Swedish population and get them to use the application in their everyday life. International Organization for Standardization’s definition of usability contains three main areas [18] :
● Effectiveness - Users achieve goals through completeness and accuracy. ● Efficiency - Provided resources have been used in relation to completeness
and accuracy through which the user has achieved the goals.
● Satisfaction - Positive attitude towards continued use of the product.
The purpose of the questionnaire is to get a deeper understanding of both the food waste problem within households but also the aspects and goals that the application had to meet in order to be used.
4.2.2 Questionnaire
The questions used in the questionnaire were kept simple, straight to the point and formed after recommendation [19]. The questionnaire consisted of 14 questions divided into four parts:
● Background of the participants ● Food waste in their households
● Fridge inventory and grocery shopping tracking in their households
● Willingness to use mobile application to track fridge inventory and grocery shopping
First part of the questionnaire generated quantitative data. The respondents answered questions regarding their age and gender, smartphone usage, household situation and if they are wasting food in their households. By understanding how often it occured we could get a sense of how big of a problem this actually is. The second part generated qualitative data. Respondents got to answer the reasons why they are wasting food in households. They answered multiple-choice questions with the most common reasons why food is being wasted but also got the chance to write their own explanations behind it. This gave evidence to the quantitative data from the previous part as well as insight into why the problem occurs. Furthermore, it led to new opportunities regarding functionality of the application. Third part generated both quantitative and qualitative data. The respondents answered if they had the problem regarding keeping track of their fridge inventory and got to either describe the problem and why or when it occurs or how they solve it. Then they proceeded to answer the following questions regarding keeping track of what to buy during grocery shopping and the occuring problems in the same way. In the fourth and last part of the questionnaire the respondents answered how willing they are to use a mobile application to keep track of fridge inventory in order to prevent food waste but also about functionality of the application itself. It generated quantitative data.
Twenty three participants in ages between 16-52 were involved in answering the questionnaire. All of them have a smartphone that they use on a daily basis and none has a background in environmental, or equivalent, studies or any knowledge about food waste, both in general and in households. However, their household situation differed.
4.2.3 Application Development
After summary and analysis of the questionnaire an Android mobile application was developed with the goal of reducing and preventing waste of fridge stored food within households. Its functionality is based on the needs and problems of consumers that emerged from the questionnaire. The study’s research question is addressed by the use of the application during a period of one-week observation. Our data collection consisted of the participants being interviewed at the end of the observation period. The interviews gathered information related to whether the users successfully reduced food wastage in their households. The result of the observation periods is analyzed based on the research questions and will additionally give insights into what needs to be improved on and what topics future research could be based on.
5. Questionnaire Result and Analysis
This section presents the results of the questionnaire.5.1 Respondents
This section presents the background information about the respondents participating in the study.
5.1.1 Age and Gender
The majority of the respondents were in the age span 18-30. Three were 30 years old or older and two respondents were younger than 18. Fifteen of the respondents were men and eight were women. See Figure 4 for age distribution and Figure 5 for gender distribution of the participants.
Figure 4: Age distribution of the participants
Figure 5: Gender distribution of the participants
5.1.2 Smartphone Usage
Respondents were asked if they own a smartphone and if so then how often they use it and how comfortable they are using it. This is important information as in order to use the application user has to have a smartphone as well as the obviousness of using the mobile phone for various purposes in their everyday life. Twenty one of the respondents use their smartphone several times a day to access other services such as BankID. The remaining two participants own a smartphone and are familiar with it however they only use it for simpler tasks such as texting and calling. Nevertheless, all respondents are owners and frequent users of smartphones.
5.1.3 Household Situation
Respondents were asked about their household situation. Fifteen lived together with at least two other people, mostly parents and/or siblings, five lived together with their partner or room mate and three lived alone.
The household situation is an important factor since it results in how much food is both purchased and wasted along with the fact that a person living alone is alone responsible for both what is in the fridge and all the waste compared with a household with several members where everyone contributes. See Figure 6 for the household situation of the respondents.
Figure 6: Household situation of the respondents
5.2 Household Food Waste
This section presents the food waste in the households of the respondents.
5.2.1 Food Waste
Respondents were asked if they are wasting food in their household. The majority, 18 of the participants (78%) responded that they are wasting food in their household and 5 (22%) responded that was not the case.
5.2.2 Food Waste Frequency
Respondents were asked about how often they waste food in their households. Six (33.3%) answered that they waste food at least once every day and eight (44.4%) that it happens one to three times a week. Four to six times a week was answered by two (11.1%), one to three times a month and four to six times a month got both answered by one (5.6%) respondent each.
It is clear that food waste in Swedish households is a major problem but also that people recognize it and are aware of their own behaviour. It is crucial that this self-awareness exists as it becomes easier to change the habits and behaviours regarding waste of food. See Figure 7 for food waste frequency of the respondents in their households.
Figure 7: Frequency of respondents food waste in their households
5.2.3 Reasons for Food Waste
Respondents were asked to answer the reasons behind the waste of food in their households. Seven (42.2%) of them answered that the reason was “Too big portion”, thirteen (68.5%) answared “Best-before-date has expired” and three (15.8%) stated that “Did not like the food”. It must be noted that similar answers, such as “Too big
portion” and “Did not finish in time” or “Best-before-date has expired” and “Food smelled / tasted bad” got categorized together.
These answers gave us that the main reason behind food being wasted in households is the fact that the best-before-date has expired. Since we expected it from previous research we could continue asking to find out why through the next follow-up question. Then followed the fact that respondents simply overestimated either their hunger or ability to eat up whatever was on their plate leading to food being wasted by throwing it away in the trash. Here, too, the respondents' self-awareness is good and they are able to answer reasons behind why the food is being wasted.
This kind of food waste is categorized as “Avoidable Waste” since the behaviour can be changed and the waste be avoided. See Figure 8 for the respondents' reasons for food waste in their households.
Figure 8: Reasons behind food waste in respondents households
5.2.4 Reasons Why “Best-before-date has expired”
Previous research has already shown that “Best-before-date has expired” is one of the most common answers to why food is being wasted. Therefore, respondents who stated that as their reason got to answer a follow-up question, namely “why did the best-before-date expire?” . Seven (53.8%) stated that they did not know how to use the food, two (15.4%) answered that they did have more of the same food at home that they used or ate instead, one (7.7%) answered that they did not like the food and the remaining three (23.1%) stated that they did not have the time to eat it yet.
More than half of the respondents’ who stated “Best-before-date has expired” as the reason behind their food waste did so because they did not know how to use the food or products while cooking. This is also “Avoidable Waste” since it can be prevented by changing behaviour. See Figure 9 for reasons why respondents answered “Best-before-date has expired”.
Figure 9: Reasons why the “Best-before-date has expired” answer was given by the
respondents.
5.3 Fridge Inventory
This section presents the fridge inventory and grocery shopping tracking of the respondents.
5.3.1 Fridge Inventory Tracking
Respondents were asked if they know and track what food they have in their fridge and if so then how do they keep track of it. Seventeen (73.9%) answered that they know their fridge inventory and 6 (26.1%) answered that they are not sure what’s inside their fridge.
Out of the seventeen, ten (58.8%) answered that they keep track of the inventory by remembering what they bought while grocery shopping, five (29.4%) answered that they keep track of it because then open it several times everyday and use or eat some of the food and the remaining two (11.8%) stated that they know it because they have a grocery shopping list on the fridge.
5.3.2 Grocery Shopping Tracking
Respondents were asked to answer how they keep track of what needs to be bought when they are grocery shopping. Figure 10 below shows their answers. Seven (30.4%) stated that they rely on their memory, seven (30.4%) that they write a list on paper and bring with them and nine (39.1%) that they write down a list on their smartphone.
Figure 10: Tracking of grocery shopping
5.3.3 Grocery Shopping Problems
Respondents were asked to answer what problems occur when they are grocery shopping. Five (21.7%) answered that they buy too much of something or something they already have at home, fourteen (60.9%) answered that they forgot to buy something and four (17.4%) had both of these above mentioned problems. Figure 11 below shows how they answered.
There is a clear problem, namely that the majority of the respondents forgets to buy something while grocery shopping even though 69.5% of them write their grocery shopping list down. Since 5.3.1, 5.3.2 and 5.3.3 directly affect each other, since you most often buy food you’ve run out of, there seems to be a big difference between how respondents experience and believe that they know their fridge inventory and thus know what they need to buy and the reality and fact that the majority forgets to buy something or buys too much of a product.
Figure 11: Respondents problems during grocery shopping
5.4 Willingness to use Mobile Application
This section presents the participants willingness to use a mobile application to help them keep track of fridge inventory and with grocery shopping.
5.4.1 Use Mobile Application to Track Fridge Inventory
The respondents were asked if they would use a mobile application in order to keep track of their fridge inventory. Sixteen (69.6%) answered that they would and seven (30.4%) answered that they would not.
5.4.2 Willingness to Take Photo of Receipt After Grocery Shopping
The respondents were asked if they would be willing to take a photo of the receipt after grocery shopping. Fifteen (62.5%) answered that they would and eight (34.8%) answered that they would not be willing to do it.5.4.3 Household’s Grocery List in the Application
The respondents were asked about their thoughts and opinions on having a joint grocery shopping list with everyone in their households. Seventeen (73.9%) answered that they would be interested in having one or at least trying it out, two (8.7%) answered that they live alone and therefore don’t have the need for it and four (17.4) of the respondents answered that they would not use or even try it. Figure 12 shows the respondent answers regarding having a joint grocery shopping list for households.
Respondents were positive to the use of a mobile application in order to track their fridge inventory and thereby reduce or prevent food waste. The majority were also willing to use the app when grocery shopping and have a shared grocery list for households, although 26.1% were not interested in using or trying it.
Figure 12: Thoughts and opinions on joint grocery shopping list for households
5.5 Conclusion of Analysis
Swedes are used to using their smartphones and various applications to different tasks in their everyday life and the participants confirmed this since the vast majority were willing to use the application described by us. The conclusions from previous research regarding both how often food is wasted in households and the main reasons behind it were confirmed. It shows what we anticipated, namely that almost all of the food wasted in households is categorized as “Avoidable Waste” meaning that by changing the behaviour of the consumers we can reduce or even prevent food being wasted.
The main reason behind why fridge stored food is being wasted is that the best-before-date has expired. More than half of the participants stated that the reason why it expired was that they did not know how to use the food while cooking which was surprising since one can assume that you buy food that you know you will consume, however that seems not to be the case. Moreover, 38.5% of the respondents who wasted their fridge stored food because of the expiry of best-before-date did so because of overconsumption while grocery shopping or lack of time. This might relate to the fact that they were not aware of when the best-before-date was expiring however it could also be a result of eating at restaurants or take-aways instead of home.
In addition, we noticed the connection between 5.3.1, 5.3.2 and 5.3.3 where not tracking food inventory, in a more methodical way than just relying on memory, results in consumers either forgetting to buy something or buying too much of something. It seems like consumers believe that they have better tracking of their food inventory than they in fact have.
In summary, in order to reduce or prevent waste of fridge stored food in households we have to change the behaviour of the consumers. By the use of our application we have to make them aware of what’s inside their fridge and most importantly when it expires, since this is the main reason why food is being wasted. Through this we can both reduce or maybe even prevent the avoidable waste of fridge stored food within households and raise concern about the topic.
6. Mobile Application Solution
The mobile application is developed for the Android operating system. Android is a free open source software that is preliminary designed for touchscreen devices where the main focus lies on its interface that is based on direct manipulation and thus it is very easy user friendly. Since it is free and open source we can implement and use several external extensions, called libraries, to develop the application based on the results of the questionnaire [20]. The application is developed using the Android Software Development Kit, coded in Java and will be accessible for all Android smartphone devices running at least the 6th released version of the operating system called “Marshmallow” [21]. Through this we will make the application accessible for people with an old and outdated Android smartphone as well as people with the latest one, in other words, we will reach the largest possible part of the population’s Android users [22].
6.1 Functionality
The functionality of the application is based and developed after the results and analysis of the questionnaire. The core functionality of the application means that the user can:
● Scan the receipt by use of mobile camera and register what’s in the fridge ● Get a notification when a product is about to expire
● Have a joint grocery shopping list with other members of the household
The three core functions and the techniques behind them are described in the following sections.
6.2 Scanning of Receipts
Users are able to scan the receipt received for the groceries they bought by the use of the mobile camera from within the application. This will be done with the help of Optical Character Recognition (OCR). The items will be added to the fridge inventory-list and users will easily be able to see and track their fridge inventory. However there is a problem, namely that the “best-before-date” is not listed on the receipt. This has been solved by setting an approximate average time that is left until the food becomes bad based on recommended calculations from the Food Marketing Institute [23]. Users are also able to set the “best-before-date” manually to match exactly what is stated on the packing of the product.
Figure 13: Scanning of receipt Figure 14: Fridge inventory list
6.2.1 OCR
The use of mobile devices for text detection and recognition in natural images are slowly becoming relevant due to the increasing amount of OCR applications and advancement in mobile technology [24]. The digitization of textual content involves several steps and OCR deals with the transformation of a digital image, for example an image of a book page in JPG format, into a digital document whose text can be processed by a computer [25]. See Figure 15 for OCRs following three logical components:
● Image scanner
● OCR software and hardware ● Output interface
The purpose of the image scanner component is to optically capture text images to be recognized. The text images are then processed with OCR software and hardware [26]. The process involves document analysis, recognizing the images and contextual processing. The output interface component allows the results to be electronically transferred into the domain that uses the results [26].
Figure 15: The structure of an OCR system [26].
There are two different approaches that we can use to implement OCR in our mobile application. The first approach would be to use an open source OCR engine. There are different types of OCR engines available and their purpose is to make it easier for developers to implement an OCR system in a mobile application. One of the most important factors that an OCR engine should hold is accuracy [27]. In order for the OCR engine to have good performance and high accuracy the engine has to be trained correctly [28]. Due to the short time-frame of the project we decided against using an OCR engine. The second approach would be to use an internet service. We decided to use TAGGUN API, which is an internet service specialized to transcribe a receipt and return relevant information [29]. This is used to parse the receipt and get hold of which groceries were bought.
6.2.2 Shelf Life of Refrigerated Foods
Previous research has shown that it is difficult to overcome the barrier of manual entry which is very user-intensive. The mobile application developed as part of this study has the functionality to manually enter the best-before date for food items. The application also uses an approximate average time for the best-before date of groceries in order to make the application less user-intensive. The approximate average time is based on calculations from the Food Marketing Institute.
Table 1: Showing the shelf life of meat and seafood [23]. Meat and Seafood Shelf Life
Bacon 7 days
Chicken or turkey 1-2 days
Chops (pork, veal, lamb) 3-5 days
Fish and shellfish 1-2 days uncooked, 3-5 days cooked
Ground meats 1-2 days
Ham, canned 6-9 months unopened, 3-5 days opened
Ham, fully cooked, half 3-5 days
Ham, fully cooked, slices 3-4 days
Ham, fully cooked, whole 7 days
Hot dogs 2 weeks unopened, 1 week opened
Lunch Meat 2 weeks unopened, 3-5 days opened
Organ meats 1-2 days
Roast 3-5 days
Sausage, hard (ex: pepperoni) 2-3 weeks
Sausage, smoked 7 days
Table 2: Showing the shelf life of dairy products.
Dairy Shelf Life
Butter 1-3 months
Buttermilk 2 weeks
Cheese, hard 6 months unopened, 3-4 weeks opened
Cheese, processed slices 1-2 months
Cheese, soft 1 week
Coffee cream 10 days
Egg Substitute 10 days unopened, 3 days opened
Heavy cream 10 days
Milk 1 week
Sour cream 1-3 weeks
Whipped cream (can) 3 months
Yogurt 1-2 weeks
Table 3: Showing the shelf life of condiment products [23].
Condiments Shelf Life
Chili sauce 6 months
Chutney 1-2 months
Cocktail sauce 6 months
Horseradish 3-4 months
Jams, Jellies 6 months after opening
Ketchup 6 months Mayonnaise 2 months Mustard 1 year Olives 2 weeks Pickles 2 weeks
Table 4: Showing the shelf life of most common leftovers [23].
Leftovers Shelf Life
Meat or poultry 3-4 days
Broth 2 days
Casseroles 3-4 days
Egg dishes 3-4 days
Fish 3-4 days
Gravy 3-4 days
Milk, condensed 4-5 days
Milk, evaporated 4-5 days
Pasta 1-2 days
Pizza 3-4 days
Salads (egg, chicken, macaroni, etc. 3-5 days
Soups and stews 3-4 days
6.3 Best-Before-Date Notifications
Users will receive a notification when a product’s “best-before-date” is about to expire. A notification is a simple message that Android displays outside of the applications user interface to show the user some information, in our case, a reminder that a fridge stored product is about to expire [30]. The application will display the notifications in three different ways as listed and presented below.
6.3.1 Status Bar and Notification Drawer Notification
When the notification is issued it first appears as an icon the status bar of the phone. The user is able to swipe down on the status bar and thereby open the notification drawer which allows them to both read more details and also take actions, such as delete the notification. The notification remains visible in the notification drawer until it is dismissed by either the application or the user [30].
Figure 16: Notification in status bar
Figure 17: Notification in notification drawer
6.3.2 Heads Up Notification
If the device is in the unlocked state the notification will appear on the top of the screen in a floating window. It appears when the application issues the notification and disappears after a couple of seconds but still remains visible in the notification drawer as usual [30].
Figure 18: Heads up notification
6.3.3 Locked Screen Notification
If the smartphone is in the locked screen mode the notification will appear on the locked screen. The detail level of the notification is set by the application. Users can also decide whether the application should show the notifications in the locked screen at all [30].
Figure 19: Locked Screen Notification
6.4 Joint Grocery Shopping List
Users will be able to create a joint grocery shopping list with other members of their households. They will be able to add, edit and remove items together with other users from the same household. This function requires the application to be connected to the internet and to a database. To achieve this the Firebase platform will be used with the Realtime Database, a cloud-hosted database, in particular.
Figure 20: Joint grocery shopping list
6.4.1 Firebase Realtime Database
The Realtime Database is a cloud-hosted NoSQL database provided on the web-based Firebase platform [31]. The data is stored as JSON and synchronized in realtime to every connected client, meaning that they all automatically receive updates with the newest data within milliseconds after the data has changed. This requires the client to be connected to the internet, however with the use of the Realtime Database the application will remain responsive even offline because the data is persisted to the disk. Once internet connection is reestablished it will synchronize itself with the current state of the server, thus also the database [31].
The Realtime Database lets us build an application with secure access to the database. Since the data is persisted locally, meaning that even when a user is currently offline, events continue to fire in realtime giving the user the best possible responsive experience. Furthermore, it is flexible regarding the security aspect giving us, the developers, total control over who has access to what data, how they can access it and how they can interact with it. Additionally, it is a NoSQL database meaning that it is only allowing operations that can be executed very quickly [31]. In summary, all of the above mentioned reasons lead to the conclusion that the use of Firebase Realtime Database is a great choice for our application and will make the user experience as smooth as possible which is crucial in order for the users to use the application.
6.5 General Data Protection Regulation
General Data Protection Regulation (GDPR) is a recent update to the data protection laws of the European Union and it significantly changed the landscape of the processing of personal data [32]. It introduced stricter requirements for the stakeholders and strengthened the rights of the individuals to whom the data are related [32]. The mobile application developed as part of this study has been tested in Sweden and since Sweden is a member of the European Union we have to take GDPR in consideration when handling the user data. When processing data there are seven principles that need to be followed [33].
1. Lawfulness, fairness and transparency — Processing must be lawful, fair, and transparent to the data subject [33].
2. Purpose limitation — You must process data for the legitimate purposes specified explicitly to the data subject when you collected it [33].
3. Data minimization — You should collect and process only as much data as absolutely necessary for the purposes specified [33].
4. Accuracy — You must keep personal data accurate and up to date [33]. 5. Storage limitation — You may only store personally identifying data for
as long as necessary for the specified purpose [33].
6. Integrity and confidentiality — Processing must be done in such a way as to ensure appropriate security, integrity, and confidentiality (e.g. by using encryption) [33].
7. Accountability — The data controller is responsible for being able to demonstrate GDPR compliance with all of these principles [33].
The mobile application was developed with these seven principles in mind. The application was built and designed in a way that every household only has one shared account. The account is created with a single email and password. The email and password is necessary for identifying and securing every user account since the household food data needs to be stored in the Firebase Realtime Database. This application design ensures that the GDPR data protection principles are followed.
7. Results
This section presents the results of the participants' use of the mobile application during a one week period.
7.1 Interview
The interviews are based on the same approach as the questionnaire, namely that they are formed after recommendation and the questions asked were kept short and simple and asking about one specific topic at a time [19]. The interview consisted of 7 questions divided into three parts:
● Household situation, usage and user experience
● Impact on fridge inventory and grocery shopping tracking ● Change of behaviour
First part of the interviews generated both quantitative and qualitative data. The respondents answered questions regarding both their household situation and how much they were using the application as well as how they experienced their interaction with it. Questions were answered both using a scale ranging from 1-5 and freely. Second part of the interview was about the application's impact on tracking food and grocery shopping. Respondents answered several questions that generated both qualitative and quantitative data. The third and last part generated qualitative data. Respondents were asked questions about their behaviour and noticeable changes in their households regarding the food waste problem.
Twelve participants in ages between 20-52 were interviewed. Seven of them were men and five were women. All of them have been using the mobile application for at least one week. None of them has a background in any environmental or equivalent studies or any knowledge about food waste or fridge and grocery shopping tracking. The major difference was their household situation.
7.2 Household Situation, Usage and User Experience
This section presents the household situation of the respondents as well as their user experience and usage of the application.
7.2.1 Household Situation
Respondents were asked about their household situation. Six lived together with at least two other people, mostly parents and/or siblings, three lived together with their partner or room mate and three lived alone. As the questionnaire showed, the household situation has a major influence on food waste since it can be directly translated into how much food is being purchased, consumed and wasted. But also the fact that a person living alone is the only one responsible for both what’s inside the fridge and what and how much food is being purchased which corresponds to the total food waste of the household.
7.2.2 Usage of Application
Respondents were asked both how often they use the application and which function they used or liked the most. Seven of the respondents used the application once a day, two several times a day and the remaining three used it 1-4 times a week. As for the most used or liked feature, eight of the respondents answered the joint grocery shopping list, two answered the tracking of the fridge inventory and two answered the receipt scanning. See Figure 21 for the respondents usage frequency and Figure 22 for the most used feature.
Figure 21: Respondents frequency of usage of application
Figure 22: Respondents mostly used or liked functions
7.2.3 User Experience
Respondents were asked about their user experience of the application. They answered both by saying a number on a scale from 1-5 where 1 was “very bad user experience”, 3 was “normal / neutral user experience” and 5 was “very good user experience” and by adding comments explaining why they answered as they did. The majority of the respondent, more precisely seven (58.3%) answered that they had a normal or neutral user experience, two (16.7%) answered that they had a good user experience, one (8.3%) that they had a very bad user experience, one (8.3%) that they had a bad user experience and one (8.3%) that they had a very good user experience. In summary, ten (83.3%) of the respondents had a neutral, good or very good user experience while using the application. See Figure 23 for the respondents answers regarding their user experience. Regarding the comments and further explanations about why the respondents answered as they did, this are some of the comments that were said:
● “The application does not look complete from a design perspective but it works really well”
● “Works flawlessly. Did not crash or interrupted once” ● “Very smooth application but it could be more polished” ● “It is a good and easy-understandable application”
Most of the comments and explanations are about the design and aesthetics of the application and although it is not directly related to the functionality it has a crucial impact regarding whether users actually return voluntarily to the application or not. Given the short time on developing the application the design was not in focus and only the standard-look of the Android components was used. It is important to notice these comments as they have a major impact in previously stated return of users. However the most important part is that ten out of twelve participants had at least a neutral and normal user experience.
This makes the application a good starting point regarding designing and developing an improved version as the core functionality is sought after and offers the users a value.
Figure 23: Respondents user experience
7.3 Impact on Fridge Inventory and Grocery Shopping
Tracking
This section presents the impact that the application had on reducing the food waste in the households of the respondents.
7.3.1 Impact on Fridge Inventory Tracking
Respondents were asked about the impact that the application had on their household food waste by tracking their fridge inventory. They answered both by saying a number on a scale from 1-5 where 1 was “Resulted in more food waste” , 3 was “Neutral / Did not notice” and 5 was “Reduced food being wasted” . Five (41.7%) aswared 4, that it had a positive impact, that is, less food was being wasted, four (33.3%) answered 3 namely that it didn’t see a difference or it did not have an impact on the food waste, two (16.7%) answered 5, that it helped them reduce the food waste and one (8.3%) stated that it had a negative impact and resulted in more food being wasted. Some of explanations were as follows:
● “It helped me realize that I need to check the best-before-dates while shopping more often”
● “Notifications are very helpful because I don’t check my fridge that often” ● “It helped us consume way more of the food we are buying rather than
throwing it in the bin”
● “I realized how monotonous and dull my diet is and how I buy food I don’t even like”
In summary, the tracking of fridge inventory had at least a neutral or positive impact on the vast majority of the respondents which correlates to elven (91.7%) of them. The comments also indicate that the fridge inventory tracking is a good functionality that helps reduce the food waste within households.
Moreover, there is a direct connection between the use of inventory tracking and the joint grocery shopping list, as the comments suggest. This shows the importance of the interaction of these two functions and strengthens the claim that the application has a positive impact on the household food waste problem.
7.3.2 Impact on Grocery Shopping Tracking
Respondents were asked about the impact that the application had on their household food waste by tracking their grocery shopping. They answered both by saying a number on a scale from 1-5 where 1 was “Resulted in more food waste” , 3 was “Neutral / Did not notice” and 5 was “Reduced food being wasted” . Six (50.0%) of them responded that they did not notice a difference, four (33.3%) that it had a positive impact and two (16.7%) that it helped them reduce the amount of food that was being wasted. For the impact see Figure 25 down below.
Some of the comments and explanations from the respondents were:
● “Joint grocery shopping list is very helpful. Could be a standalone application”
● “This and fridge tracking help both remember what to buy and reduce food waste by not overconsuming “
● “I did not notice a difference in food waste but it is very helpful and easy to use”
● “Helped me only buy what is really necessary and not buy too much of food that I already have in the fridge”
Half of the respondents stated that grocery shopping tracking had a positive impact on the food waste in their household, along with the fact that no one of the respondents answered that it had a negative impact on the food waste. As stated in 7.3.1, the cooperation between the fridge tracking and grocery shopping tracking is received very well by the users. This reinforces the motivation that a future improved version of the app should contain both features, as they work and have a positive impact on the reduction of food waste both individually and when cooperating.
Figure 25: Grocery shopping trackings impact on respondents household food waste
7.4 Change of Behaviour
This section presents the impact the application had on the change of behaviour of the respondents regarding their household food waste.
7.4.1 Noticable Change of Behaviour
Respondents were asked about the impact that the application had on their behaviour regarding household food waste and if they noticed any changes or differences compared to when not using the application and answered freely. The first question asked was “Did you notice a change in your behaviour regarding the food waste in your household while or after the one week period?” . Nine (75%) of the respondents stated that they did notice a positive change of behaviour and three (25%) that they did not notice a change of behaviour. None answered that they noticed a negative change of behaviour. See Figure 26 for respondents answers. Further follows some comments and explanations regarding the impact of the application on the change of behaviour:
● “It made me think twice about what I need to buy while grocery shopping” ● “Application showed me how much food (and money) my household is
unnecessarily wasting”
● “I did not know that I spend so much on food I don’t even consume”
● “This made me realize that I don’t have to buy lunch in restaurant since I have too much food at home that I am wasting”
● “Grocery shopping became much easier and convenient. Don’t have to call or text my partner since he can just check the application directly”
Although all comments were not pointing directly to the reduction of the food waste problem they made it clear that the use of the application had an impact on the change of the respondents behaviours. A more detailed and in-depth analysis will be found in the following main heading.
Figure 26: Respondents answers regarding noticable change of behaviour
7.4.2 Reasons Behind Change of Behaviour
The second question the respondents were asked was “When did you notice the change of behaviour?” . This question was only asked to the nine respondents who answered that they noticed a change of behaviour in the previous question. Five (55.6%) responded that it was during or after the use of the grocery shopping-functionality, two (22.2%) that it was during or after the use of the fridge inventory tracking and two (22.2%) that it was while throwing away food in their household. See Figure 27 below for the reasons when respondents noticed the change of behaviour. The most common explanations and comments were:
● “While grocery shopping I realized that I did not know if I had several products at home or not so by checking my fridge inventory-list I could quickly see what needed to be bought and what I already have”
● “I was throwing away food and it hit me that it could easily have been prevented if I just used and checked the application more frequently”
● “Was buying groceries and while checking the list I could see that a new item has been added by my wife so I could instantly go and add it to the basket and buy it”
It is clear that the majority of the respondents realized a change in their behaviour during or after using the grocery shopping tracking function. It seems that this function is the one that is affecting people the most. A more detailed and in-depth analysis will be found in the following main heading.
Figure 27: Respondents answers regarding when they noticed the change of
behaviour