Reduction of food waste in the Dutch catering sector | Author: Han Soethoudt [ Auteur: Han Soethoudt ] | Report no. 1335 |

Title | Reduction of food waste in the catering sector | Author | Han Soethoudt | Number | Food and Biobased Research number 1335 | ISBN-number | ISBN no. 978-94-6173-336-8 | Date of publication | Date of publication August 2012 | Confidential | No | OPD-code | 6239021000 | Approved by | Toine Timmermans | | Wageningen UR Food & Biobased Research | P.O. Box 17 | NL-6700 AA Wageningen | Tel: +31 (0)317 480 084 | E-mail: info.fbr@wur.nl | Internet: www.wur.nl | | © Wageningen UR Food & Biobased Research, institute within the corporation Stichting Dienst Landbouwkundig Onderzoek (Foundation Service Agricultural Investigation) | Alle rechten voorbehouden. Niets uit deze uitgave mag worden verveelvoudigd, opgeslagen in een geautomatiseerd gegevensbestand of openbaar gemaakt in enige vorm of op enige wijze, hetzij elektronisch, hetzij mechanisch, door fotokopieën, opnamen of enige andere manier, zonder voorafgaande schriftelijke toestemming van de uitgever. De uitgever aanvaardt geen aansprakelijkheid voor eventuele fouten of onvolkomenheden.All rights reserved. No part of this publication may be reproduced, stored in a retrieval system of any nature, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher. The publisher does not accept any liability for inaccuracies in this report. | |

Colophon

Project partners

This study has been made possible with co-financing as innovative pilot by the Platform
Verduurzaming Voedsel (Platform Preserved Foods) and by the Interreg IVB project GreenCook

Summary

Nowadays, almost every stakeholder in and around the food chain is paying attention to durability. One example of this durability is the theme food waste which is a familiar subject to many people. It could generate money, improve the image of companies and is high on the durability agenda of the government. As a result, many initiatives have started up with regard to this theme, not only in the Netherlands but especially in the United Kingdom.
Apparently, we do not know how many food is wasted exactly and it is mainly based on rough estimations on a higher level of abstraction. This makes it difficult to take specific measures and reduce the amount of waste.

In 2010, Veneca decided to take the initiative to tackle this problem which has led to a unique cooperation between all members of this umbrella organisation that operate in a highly competitive market. A similar open cooperation between companies had not taken place before in the area of food waste in the Netherlands. On the one hand, trust was offered in carefully dealing with information about waste and on the other hand, full cooperation was provided to a very extensive and thorough analysis of many locations in this area. Due to these activities, a very reliable and detailed investigation has taken place.

The results can be divided into the amount of food waste on a product level, causes of that waste and testing an improvement measure in order to reduce food waste. Based on previous investigation, we only focussed on what had to be thrown away of the lunch based on the food that was offered. The quantity that remains as a result of the preparation by the catering manager and the leftovers on the plate of the consumer is a relatively small amount of waste.
The waste in de catering sector, except for the institutional and inflight catering, which each have different and very specific logistics and dynamics, amounts to approx. 104 tons a week, i.e. 5 million kg a year. In perspective of an estimation for food waste and –loss, between 8 and 9 billion kg in the Netherlands, this is not so much.
The most important contribution in weight for waste originates from soup and the salad bar with dressing which both represent almost 50% of the weight. It is true that soup mainly consists of amount of water but it is costly with regard to the CO2 equivalents during preparation, as soup has to be kept warm for a rather long period in time.

Extensive research has been performed to find causes of waste on the catering location. Most important reason for waste is the fact that you do not know the exact number of guests. This reason is important with regard to the amount of food that has to be prepared and taken out of the freezer. In many cases, it is not possible to bring out products on demand because you can not keep your customers waiting too long. These insights enable caterers to anticipate on waste on a product level.

It is nevertheless evident that the estimation of the number of guests has a lot of impact on the potential amount of reduction of food waste. For this reason, this measure was tested in a small pilot.
This measure has proven to be complex because cooperation with the customer is a required condition. In addition to this, it proved not easy for catering managers to prepare the required amount of food even if they knew how many guests were expected. However, there are indicative results that suggest that knowing the amount of guests is very useful. With regard to the relevance of it for all products involved, attention for this measure is strongly recommended as this contains a possibility to reduce a considerable amount of food waste. Other options to reduce waste in catering are flexibility with regard to the obligations in the contract and/or a menu cycle based on local consumer preferences, a carefully dosed adding of products during the lunch period, preparing food on demand at the end of the lunch and adding economic value to the product by integrating it in another product or bringing it to a high-quality flow of cattle feeding instead of wasting it.

Table of contents

1 Introduction 6
2 Scope and objective 7
3 Approach 9
4 Results 12
4.1.1 Potatoes, vegetables, fruit 12
4.1.2 Bread 14
4.1.3 Soup 16
4.1.4 Fish 17
4.1.5 Meat 18
4.1.6 Dairy 19
5 Conclusions and recommendations 29
6 Acknowledgement 31
Literature 32 Enclosure A Interview table product category 33 Enclosure B Example of filled in table of a catering location 35 Enclosure C Results workshop with regard to improvement measures to reduce food waste 36 Enclosure D Food waste survey in catering sector 37 1 Introduction

Since a few years, durability is a socially accepted theme which is supported by civilians and society as a whole. Most stakeholders realize the importance of this theme and show initiatives in this area or would like to do so. Within durability as a theme, food waste is an important topic. In June 2010, the CBL has appointed durability as most important and first issue to be tackled. The LNV has assigned priority to this topic as well by developing a monitor on this theme. For members of Veneca food waste is a relevant subject as well deserving proper attention and that is the reason we would like to start up a common project. This willingness to do so is very special because both parties are able to distinguish themselves in the field of durability. All the same, all parties involved would like to take on food waste together under the Veneca flag.

Little is known about the quantity of food that is wasted in catering. In the “factsheet Food Waste of May 2010” of LNV, it is mentioned that in retail and out-of-home market 3-6% is wasted, while another source ([1]) reveals that 1-5% of prepacked food and sometimes up to 40% of unpacked products are wasted in the out-of-home market. If you would like to tackle such a problem, it is essential to look at the place where the largest amount of food is wasted and if that is not known, further investigation is required. It should be noted that waste might not be product-related but process-related or a combination of these two, for example “self-prepared desserts”. Later on, this will be elaborated on in the approach of the project.

2 Scope and objective

In literature, a distinction is always made between avoidable and unavoidable food waste. Avoidable food waste are for instance filled sandwiches that remain after lunch, unavoidable food waste are animal bones and orange peels. In this project, only the avoidable food waste will be focussed on. In addition, everything is included what the caterer has not sold and thus has to throw away. This means that leftovers on the plates of people eating lunch are not included as they have already paid for this food. Reuse for animal foods, food pantries or other purposes is not included either.
Catering is a wide-ranging activity and it is not possible to include everything, i.e. schools, nursing homes, inflight caterers, party caterers etc. That is the reason why a definition has been formulated. In this project, catering is mainly focussed on, which is covering a major part of the sector on the one hand and has a similar logistical structure on the other hand. This means that company catering at companies, schools and public bodies are included but nursing homes, hospitals, party catering and inflight catering are excluded. Serving the consumer/patient can not be compared in these situations, as a result of which the improvement measures will not (easily) be transferable. These types of catering require another and more specific approach.
Another definition is the fact that only 50% of the catering has been outsourced in the Netherlands. Approx. 90% is executed by Veneca members. The division over the sub sectors looks like the below-mentioned table (see [4]):

2010 Catering | Turnover (million Euros) | Turnover (%) | Companies | 926 | 84.6% | Institutions | 85 | 7.8% | Education | 83 | 7.6% | TOTAL | 1,094 | 100% |
Table 1: turnover catering Veneca members 2010 (inflight is included under companies)

This part of the investigation focussed on that part of the catering that was covered by the Veneca members via companies and education. As KLM Catering Services is a member of Veneca as well, for this organisation a (limited) effort was made to consider the options to reduce food waste.

It is important to elaborate further on the way in which facts were measured. The following choices were made. The investigators decided to measure in kilogrammes and not in percentages and money. Measuring in kilogrammes is relatively easy (despite the considerable effort that has to be made as well). What is more, it can be added up and compared quite easily.

That is not the case if they would have measured in a certain amount of money. First of all, you have to deal with sales prices which are different everywhere and depend on the type of contract. In some cases, the milk is free of charge and in other cases, they charge € 1.25. The result would have no value because it can not be compared with other sectors where prices are more similar.
A better choice would be the purchase value which involves other obstacles. The information is confidential and can not be shared and what is more, the result can not be compared with anything.
With regard to the percentage, a practical consideration has been made. Weighing, writing down and sending data was already a heavy task for the catering managers. It would be asking too much of them if we had also asked them to weigh the amount of food they had prepared (including the preparation between times). The consequence of this is that absolute quantities in kg should be put into the proper perspective. It might be possible that a lot of a product is wasted while it amounts to only 2% of the offered quantity for example and the catering manager has actually made a correct assessment.

With this definition, the objective can be formulated as follows:

Objective:Properly mapping the largest avoidable food waste (in kg) in the company- and educational catering sector and finding solutions for the entire sector, including the evaluation of these solutions by means of pilots. In the process of finding solutions, prioritizing is also based on CO2 emission (or equivalents). |

3 Approach

The catering market is wide-ranging. On a regular basis, it is claimed that every location is unique. In order to attain the objective, an extensive and thorough investigation is therefore required. For this, a sound balance is required between representativeness on the one hand and the amount of work on the other hand.

Phase I: measuring food waste
In the Netherlands, over 4,000 are classed under Veneca members of which over 300 institutionally. As approx. 50% of the market has been outsourced, we are able to conclude that an investigation that is representative for Veneca, would be representative for the catering sector as well (obviously taking the formulated definition into account, for example the exclusion of certain organisations).

Step 1: Segmentation
To avoid that thousands of locations are involved in the investigation, use has been made of a segmentation, which has been achieved in consultation with the Veneca members. This segmentation is based on the most important dimensions on which the locations differ from each other:

• Type of location school (1), company (2), other (3)
• Type of contract open book (1), with commercial elements (2)
• # lunch users per day < 100 (1), 100-250 (2), > 250 (3)
• Meal types lunch (1), lunch + dinner (2)

As a result, the total amount of combinations is 3 x 2 x 3 x 2 = 36 classes of catering locations.

Step 2: Selection of locations
Per class, members of Veneca were asked to list locations in order to have a representative amount in their sample survey of all common classes. In the end, 8 caterers (Albron, Appel, Avenance, Compass, Hutten, Prorest, SAB and Sodexo) mentioned 25 locations and on 200 locations, food was weighed.

Step 3: Linking product groups and caterer
In order to avoid overburdening the catering manager, it was decided to weigh the loss of 2 or 3 product groups per caterer. The following product groups were selected: potatoes, vegetables, fruit, bread, soup, fish, meat and dairy. This selection has been established after careful consideration, because these groups will probably cause the most loss. In addition to this, a list with products was established per product group representing the major part of that product group in terms of turnover. An example of this can be found in Enclosure A.

Crucial in this investigation is measuring on a product level, so that is becomes clear where and how matters can be changed to reduce food waste.

Step 4: Interviews with all 200 locations
These 200 locations have been interviewed about all types of location-specific matters such as range of products, food offering frequency, delivery frequency, type of presentation, packaging, the way food is added during lunch etc. An example of this can be found in enclosure A. The first 25 locations were visited personally, while the other 175 were contacted over the phone (approx. 30-45 minutes per interview). The required information is used for calculation on the one hand and retrieving causes of food loss on the other hand.

Step 5: Measurements
The measurements were executed over a period of two weeks. Measurements were recorded in grammes. Based on the weighing instructions on a product level, the first weighing was always performed by project assistants or by temporary workers. This was a deliberate choice to obtain a reliable measure of reference and a clear transmission of the measuring instructions. The other measurements were executed by employees of the catering organisation. An example of a completely filled in table can be found in Enclosure B.

Step 6: Data processing
If the pilot period is terminated, the information will be joined together and scaled up. To this end, classifications were retrieved at all caterers for every segment as described in step 1.

Phase II: determination of causes
Based on the interviews and an Internet survey, the most important causes of food waste are determined and improvement measures identified for short, medium length and long-term periods.

Step 7: Analysis causes food waste
The combination of weighing results and circumstances on that location provide insight into possible causes based on correlations. In addition to these 200 locations, all other locations of the Veneca members were included in the survey with questions about their location and their opinion about the causes of food waste. The survey was distributed among 3,500 locations. The catering managers were asked what the most important cause of waste was for every product on their location.

Phase III: identification of improvement measures with regard to food waste

Step 8: Brainstorm session with experts of the caterers involved

In a workshop, organised by Wageningen UR, company experts were involved methodically in finding and prioritizing the use and feasibility of solution approaches to reduce food waste on the catering locations. Solutions had to be identified for short, medium length and long-term periods.

Phase IV: testing improvement measures in pilots to reduce food waste

Step 9: Pilot
The outcome of phase II will determine the interpretation of the pilot. Taking the scope of phase I into account, there is only limited space to perform the pilots. That is the reason why it was decided to test one improvement measure per caterer on one location.

Step 10: Analysis of pilot
The data with regard to food waste will be analysed before and during the implementation of the improvement measure.

4 Results

The results can be divided into four parts. The first part contains a summary of the first two phases as this will enhance the readability. The measuring results of the measurements of loss can then be linked directly to the causes. In paragraph 3 and finally in paragraph 4, the outcome of the pilots is described in which improvement measures were tested to reduce food loss.

4.1 Phase I and II

The results of the food waste measurements will be presented per product group first. Then they will be incorporated in a total overview and in certain sections of location criteria, for example school vs. company vs. other. As already mentioned, the approach for the categories that were investigated, the most common products, representing the category excellent, the amounts of waste were weighed. As part of durability, the CO2 equivalents in kg were considered (where possible).

Part 1: food waste linked to product (group)s 4.2.1 Potatoes, vegetables, fruit

In the category potatoes, vegetables, fruit, it is obvious what causes waste. Across the entire sector, the most waste is caused in salad bars. Naturally, there is a simple explanation for this. What is offered most of potatoes, vegetables and fruit is also present in these two “products” despite the fact that a salad bar is not present on all locations.

Potatoes, vegetables, fruit | Kg/wk | % potatoes, vegetables, fruit | Salad bar with dressing | 12,540 | 48.2% | Salad bar without dressing | 7,858 | 30.2% | On sandwich (potatoes, vegetables, fruit) | 1,325 | 5.1% | Hot meal (potatoes, vegetables, fruit) | 1,260 | 4.8% | Fruit salad | 1,056 | 4.1% | Fresh fruit | 868 | 3.3% | Other | 585 | 2.3% | In bowl without dressing | 392 | 1.5% | Cold snack | 109 | 0.4% | In bowl with dressing | 0 | 0.0% | Total | 25,993 | 100.0% |
Table 2: food waste on product level in category potatoes, vegetables, fruit in the catering sector
However, there are more causes. The following table is depicting the results of the Internet survey. The answers are different per category and left vacant (i.e. not selected from a list of options).
The complete list with answers is, where possible, joined together so that approx. 20 causes remained.

Figure 1: causes of food waste per potatoes, vegetables and fruit product in accordance with catering managers
To illustrate the figure, a graphic point is selected:

Product: fresh fruit Cause: quality Value: 28%

The meaning of the above-mentioned is the fact that 28% of the catering managers with fresh fruit in their range of products and suffering from waste as a result, indicate that this is caused by the quality of the fruit during delivery.

Note: one specific “cause” was mentioned quite often and has therefore been included in the graphic: “soup left”. Naturally, this is unusable information.

By transposing the data (turning the axis), insight into the matter is increased and it becomes more obvious which causes really matter regarding the products. Due to the complexity of such graphs, the most important products regarding food waste (the salad bar with and without dressing) are depicted in the graph. The other product categories will be depicted like this as well. This way, it becomes quite clear as well that “very difficult to estimate the number of guests” is the most important cause.

Figure 2: causes of waste at a salad bar with and without dressing and potatoes, vegetables and fruit on a sandwich
By analysing the products that contribute the most to the food waste this way, it becomes obvious in which way waste could be reduced.
Conclusions of potatoes, vegetables and fruit: * The cause “amount of guests difficult to estimate” is the most important one in the category potatoes, vegetables and fruit * Quality during delivery of fresh fruit is an important cause of waste * A number of products are not popular with the consumer

4.2.2 Bread
Bread is mostly wasted in the category separate slices of bread, soft white rolls, luxury rolls and filled sandwiches.

Bread | Kg/wk | % bread | Separate slices of bread, soft white rolls | 6,394 | 41.4% | Baked rolls | 3,985 | 25.8% | Filled sandwiches | 3,092 | 20.0% | Baker baked bread | 833 | 5.4% | Large loaf of bread, baked yourself | 753 | 4.9% | Large load of bread, baked by baker | 389 | 2.5% | Other (bread) | 0 | 0.0% | TOTAL | 15,445 | 100.0% |
Table 3: food waste on a product level in the category bread in the catering sector
These products together represent 80% of the waste in this category, which is the consequence of the fact that these products are offered on most places.
In figure 3, it can be detected that these products are sold in 90% of the investigated locations and less than 50% on other locations. It should be noted that being present in the range of products is not sufficient because it is possible that the product is only offered once a week (for example fish). This has been taken into account. However, the offering frequency was daily for the most part, so that the difference was small in the weighing process.

Figure 3: share in % of the 200 measuring locations where a certain product is included in the range of products If we look at causes for the most important products, we can see the result in figure 4.
It is striking that the causes do not differ so much per product and the most important cause again is “difficult to estimate the amount of guests”.

Figure 4: causes of waste of products in the category bread that contribute the most to waste

Conclusions bread: * The cause “amount of guests difficult to estimate” is also the most important one in the category bread * There is a variety in the types of bread, sandwich filling and luxury rolls and in many cases, it is not clear what the preference of consumers is (often the same group of people) * In case of bread that must be finished off in the oven, one does not want to keep the customer waiting and bread is already finished off in the oven. As a result, the estimation of the catering manager is decisive for the amount of waste. |

4.2.3 Soup
Soup is a convenient category because in most cases, soup is offered in large pans (86% of the 200 locations) and/or in bowls (16%). Soup seems to be the category in which the most is wasted in kg. The waste in bowls is relatively small.

Soup | Kg/wk | % soup | Large pan self-service | 34,446 | 95.1% | In bowls | 1,758 | 4.9% | TOTAL | 36,205 | 100.0% |
Table 4: food waste and CO2 equivalents of soup in the catering sector
From the table, it can be concluded that the CO2 equivalents are very high. Ultimately, soup has the highest score of all products. A large amount of soup consists of water and that is why it is surprising that so many CO2 equivalents are involved. This can be explained by the fact that soup is heated early in the morning up until lunch, as a result of which a lot of energy is used.
In the answers with regard to the causes, the catering managers made no clear distinction between causes for soup in large pans on the one hand and soup in bowls on the other hand.

Figure 5: causes of waste in soup

It has become clear that the large variety of soup (for instance due to a menu cycle) also means that some soups are more popular than others. On a regular basis, people have proposed to decrease the variation in soups based on the experiences the catering manager has with the offered soups.

Conclusions soup: * Soup is a category in which much is wasted (in kg) * The corresponding CO2 equivalents are very high with regard to this waste * De cause “amount of guests difficult to estimate” is the most important cause in the category soup * Catering managers regularly mention that people eat less soup if it is warm. * The menu cycle is one of the causes that the soups offered are less popular |

4.2.4 Fish
The category fish deviates from the other categories as the offering frequency of it is much lower. The average frequency amounted to less than once a week on 200 locations.

Fish | Kg/wk | % fish | Fried fish (cod parings) | 237 | 45.5% | Fish on sandwich (salmon) | 221 | 42.3% | Ready-to-eat cold fish | 64 | 12.2% | Other (fish snacks) | 0 | 0.0% | TOTAL | 522 | 100.0% |
Table 5 : food waste of fish in the catering sector
Across the sector, that is 424 kg/week corresponding with an average of 120 grammes per week per location. The largest waste products “fried fish” and “fish on sandwich” causes are mentioned in Table 5.

Figure 6: causes of waste in fish
Several causes can be detected here. “Some of the products are not popular”, “the number of guests is difficult to estimate” and “difficult to predict if they are in the mood for this product” stand out.

Conclusions fish: * Fish is a category in which little is wasted per kg * Most important reason for this is that it is offered in small quantities with an average of 1x per week * The next 3 causes contribute equally to waste: “some products are not popular”, “the number of guests is difficult to estimate” and “difficult to predict if they are in the mood for this product” | 4.2.5 Meat
The category meat is present in many products during lunch: soup, snack, on sandwich, hot meal and salad. Meat produces a large amount of CO2 equivalents, so a high environmental pressure, as illustrated in the below-mentioned table. The weight of waste will be multiplied by ten in the CO2 equivalents where most categories are less in weight.

Meat | kg/wk | % meat | Meat snack | 2,776 | 38.9% | Assorted sliced cold meat (not packaged, separately) | 1,779 | 24.9% | Sliced cold meat (packaged) | 843 | 11.8% | On sandwich (meat) | 662 | 9.3% | Added in soup | 567 | 7.9% | Hot meal (meat) | 420 | 5.9% | Added in salad | 89 | 1.2% | Other (meat) | 0 | 0.0% | TOTAL | 7,136 | 100.0% |
Table 6: food waste and CO2 equivalents of meat in the catering sector
The snack and the unpacked meat products are focussed on to investigate the causes:

Figure 7: causes of waste in meat

It is striking that for the meat snack something typical has happened for deep-frozen products, namely that too much was taken out (this is also common for fried fish). As the customer should not be kept waiting too long the deep-frozen products are taken out of the freezer in advance. The exact amount if difficult to predict. The “amount of guests is difficult to estimate” is once again the most significant reason. If meat products are regarded separately, then it is not possible to assign an (apparent) cause.

Conclusions meat: * Most waste is created with meat in “meat snacks” and the “separate meat products” * Most important reasons for meat snacks are the unknown number of guests and the fact that the product has to be taken out of the freezer, so that you have to make an estimation of the quantity in advance. * Meat contributes considerably to the CO2 equivalents and in an absolute sense, the environmental pressure in this category is very high (second after soup). | 4.2.6 Dairy
In the category dairy, the (mainly simple) products with the least environmental pressure produce the most waste.

Dairy | Kg/wk | % dairy | (Butter)milk in jar | 8,021 | 44.8% | Boiled eggs | 1,451 | 13.7% | Added value dairy drink | 1,931 | 10.8% | (Butter)milk in dispenser | 1,146 | 6.4% | Margarine and butter tubs | 969 | 5.4% | (Butter)milk in packages | 948 | 5.3% | Packaged dessert | 914 | 5.1% | Self-made dessert | 676 | 3.8% | Cheese (separate slices) | 356 | 2.0% | Snack with dairy | 176 | 1.0% | Cheese on sandwich | 158 | 0.9% | Fried eggs | 119 | 0.7% | Packaged cheese | 26 | 0.1% | Other (dairy) | 0 | 100% | TOTAL | 17,891 | 100.0% |
Table 7: food waste and CO2 equivalents of dairy in the catering sector

The cheese products have a low waste but a common environmental pressure resembles products as of the third place in the list. For two products with a lot of waste, the “use before” date ( too limited) was indicated as the reason for waste, while the unknown amount of guests was not playing such an important role here.

Figure 8: causes of loss in dairy Conclusions dairy: * In this category, ordinary milk (in a jar and in packages) is the product with the most waste * The catering managers indicate that the “use before date” was too limited and because of that, the most important cause |

If the “use before date” is too limited, this has to do with a number of factors such as minimal order quantity versus rate of circulation, delivery frequency and order behaviour. Every caterer who would like to know more about the cause, will have to look at the situation on their own locations.

Part 2: food waste, total survey of all products
After joining all products, the following survey of food waste is created: | Category | Product | Kg/wk | % of total | Cum% | CO2 (kg)/kg | CO2 (kg)/wk | 1 | Soup | large pan self-service | 34,446 | 33.2% | 33.2% | 3 | 103,339 | 2 | Potatoes, vegetables and fruit | salad bar with dressing | 12,540 | 12.1% | 45.3% | 1.2 | 15,048 | 3 | Dairy | butter)milk in jar | 8,021 | 7.7% | 53.0% | 1.6 | 12,833 | 4 | Potatoes, vegetables and fruit | salad bar without dressing | 7,858 | 7.6% | 60.6% | 0.5 | 3,929 | 5 | Bread | separate slices, soft white rolls | 6,394 | 6.2% | 66.7% | 0.7 | 4476 | 6 | Bread | Luxury rolls (croissants, hard crusted rolls) baked yourself | 3,985 | 3.8% | 70.6% | 0.7 | 2,789 | 7 | Bread | filled sandwiches | 3,092 | 3.0% | 73.6% | 0.7 | 2,164 | 8 | Meat | meat snack | 2,776 | 2.7% | 76.2% | 9.9 | 27,486 | 9 | Dairy | Boiled eggs | 2,452 | 2.4% | 78.6% | 1.4 | 3,433 | 10 | Potatoes, vegetables and fruit, 2/3 and meat 1/3 | on sandwich | 1,987 | 1.9% | 80.5% | | |
Table 8: food waste and CO2 equivalents of dairy in the catering sector

For a complete table, we refer to enclosure D. In total, approx. 104 tonnes of food is wasted per week in the catering sector. This amounts to approx. 5,000 tonnes annually. In the Netherlands, 8 to 9 million tonnes of food is wasted each year (avoidable and unavoidable) which means that the catering sector (outsourced and under own management) contributes approx. 0.1% to this.
Soup is the largest cause of waste but the salad bar is a good second. About 10 products cause 80% of the weight in food waste.

Part 3: food waste in different sections of categories on locations
The approach of this measurement is based on a classification that was agreed upon, described in chapter 3. The result is an addition of the waste of all types of locations. The numbers mentioned by the caterers per type is taken into account. It is therefore useful to take a step in order to see if there are differences within these dimensions. The following dimensions were used:

* Type of location: school (1), company (2), other (3) * Form of contract open book (1), with commercial elements (2) * Amount of people using lunch per day < 100 (1), 100-250 (2), > 250 (3) * Meal types lunch (1), lunch + dinner (2)

With regard to the first three, the differences are significant while for the last one, the number of locations with dinner was relatively low. As a consequence, the result was not significant enough to make a distinction. The results of the first three dimensions are mentioned below. Type of location: school, company, other

From the amounts per type the caterers have delivered, the division of these three are as follows:

school | company | other | 3.5% | 58.2% | 38.4% |

The division of food waste on a product level for the most important products is mentioned in the table below: Table 9: division of food waste per product with regard to the three types of location

In the table, a distinction has been made between the waste within the type and between the types. A large pan of soup in the self-serving section is responsible for 29.6% of the food waste within school catering and only over 9% in the category “other”. In addition to this, the large pan in the self-service section at school is only responsible for 2.5% of the waste in the entire sector. It is important to look at this in the correct perspective. The number above the type (green headers) are related to the amount of locations in that category. If only 3.5% of the locations are schools, then it is logical that the contribution in waste is not substantial anywhere. Soup is with its 91.7% a typical phenomenon in companies.
Type of contract: open book, with commercial elements

The division that can be made based on the information of the caterer is: Open book | Commercial element | 43.4% | 56.6% |

The division of waste of the most important products is mentioned in Table 10

| % within type of contract form | % across type of contract | Product | open book | comm. elem | open book | comm elem. | Large pan self-service | 39.06% | 29.7% | 43.9% | 56.1% | Salad bar with dressing | 8.7% | 14.1% | 26.8% | 73.2% | (Butter)milk in jar | 7.5% | 7.9% | 36.1% | 63.9% | Salad bar without dressing | 5.8% | 8.6% | 28.5% | 71.5% | Separate slices, soft white rolls | 5.7% | 6.4% | 34.8% | 65.2% | Luxury rolls (croissants, hard crusted rolls) baked yourself | 2.0% | 5.0% | 19.2% | 80.8% | Filled sandwiches | 1.0% | 4.2% | 12.2% | 87.8% | Meat snack | 3.4% | 2.2% | 47.4% | 52.6% | Boiled eggs | 3.2% | 1.9% | 50.6% | 49.4% | On sandwich | 0.9% | 2.5% | 18.2% | 81.8% | Added value dairy drink | 0.8% | 2.5% | 15.9% | 84.1% | Meat unpacked separate | 4.0% | 0.3% | 87.3% | 12.7% | Soup in bowls | 4.59% | 0.3% | 100.0% | 0.0% | Cum % | 86.5% | 85.3% | | |
Table 10: division of food waste per product with regard to type of contract

Not many striking differences here, other than the relationship between waste of several bread products between 10-20% while the type of contract has another proportion. It is possible that part of the solution can be found in the range of products offered.

Please note: the product “on sandwich” consists for 2/3 of potatoes, vegetables and fruit and for 1/3 of meat waste.

Amount of people using lunch per location: <100, 100-250, > 250

The division of these classes is as follows: small | medium | large | 59.5% | 28.1% | 12.3 % |

The following table shows us that within the classes, little difference can be detected. Across the types, significant deviations are noticed. In particular the added value of dairy drink stands out, as in the smaller locations (in round figures), nothing is wasted. In addition to this, the waste of (butter)milk in a jar is substantial with 46.6% on the large locations.

| % within size type of location | % across size type of location | Product | small | medium | large | small | medium | large | Large pan self-service | 32.4% | 30.4% | 38.3% | 40.0% | 31.3% | 28.7% | Salad bar with dressing | 9.9% | 15.8% | 10.6% | 33.7% | 44.5% | 21.8% | (Butter)milk in jar | 5.2% | 5.8% | 14.5% | 27.8% | 25.6% | 46.6% | Salad bar without dressing | 8.0% | 6.2% | 8.7% | 43.6% | 27.9% | 28.5% | Separate slices, soft white rolls | 9.5% | 4.0% | 3.7% | 63.2% | 22.0% | 14.8% | Luxury rolls (croissants, hard crusted rolls) baked yourself | 6.6% | 1.5% | 2.5% | 70.2% | 13.7% | 16.1% | Filled sandwiches | 2.7% | 3.7% | 2.4% | 37.6% | 42.2% | 20.3% | Meat snack | 2.3% | 2.1% | 4.1% | 34.8% | 27.0% | 38.2% | Boiled eggs | 2.7% | 2.2% | 1.9% | 47.7% | 32.4% | 19.9% | On sandwich | 1.2% | 2.9% | 1.8% | 25.2% | 52.1% | 22.7% | Added value dairy drink | 0.0% | 5.2% | 0.3% | 0.0% | 96.0% | 4.0% | Cum % | 80.6% | 79.8% | 88.8% | | | |
Table 11: division of food waste per product with regard to size of location

It is quite evident that conclusions should be drawn with some caution. The contribution to waste could be substantial somewhere just because this product is offered in larger quantities on this location or part of the range of products more often within that category. It is not fair to say that one category location is handling this better than another location. This would require a more profound analysis.

What can be deviated from this information, is the fact that somewhere, a significant quantity is wasted. It would be advisable to act upon this. A clear example of this is the added value of dairy drink at companies.

4.2 Phase III

In this phase, improvement measures have been listed by means of a workshop with experts on behalf of the caterers. The detailed results are mentioned in Enclosure C. In table 12, they are listed per subject in a compact form.

| | Keeping score: estimated chance for success | Category | | high | medium | low | A | Better insight in amount of visitors | 6 | 6 | 0 | B | Adaptation of product range | 11 | 6 | 1 | C | Adapting method of working on location | 19 | 7 | 0 | D | Issues related to the distributor | 0 | 4 | 1 | E | Consumer | 5 | 9 | 5 | F | Matters regarding the customer | 5 | 7 | 0 | G | Matters regarding catering personnel | 2 | 5 | 3 | H | Process-related/technical | 8 | 1 | 4 | I | Miscellaneous | 0 | 3 | 2 |
Table 12: expert scores on main directions in which improvement measures are considered to be realistic.

All measures put together that are included in the category range of products adaptation, the experts have indicated 11 times that they expect to be able to achieve much with this. For changing the method of working on the location, this even amounts to 19 times. Such a schedule has also been drawn up for the specific measures (within a category from the above-mentioned schedule) which results into the following table:

Category | Top 5 high | Score | A | Customers should inform us about special occasions (prior to incentive) | 5 | C | Add products during process | 4 | B | More madiwodo-soup/sandwich | 3 | C | Gone = gone | 3 | C | Smaller containers at the end | 3 | H | Residual processing in menu planning | 3 |
Table 13: list with best scoring improvement measures according to experts

The term “madiwodo-soup” (soup from Monday, Tuesday, Wednesday, Thursday) and madiwodo-bread refers to soup and bread on which food is used that was left over last time.
The most important improvement measure refers to the indication of a possible variation in the amount of guests that can be expected for lunch. The appealing aspect of this measure is the fact that they are not product-specific and therefore helpful to all products. This measure is applicable to the outcome of paragraph 4.1 in which “the unknown number of guests” is the most important cause of waste by far.

4.3 Phase IV

In the last phase of the project, it was possible to test one improvement measure in a pilot. Based on the previous analysis, it was emphasized that the effect of knowing how many guests will arrive should be investigated with regard to the amount of food waste. This measure is complex because of a number of reasons.

a) Acquiring this data has to be arranged via the customer, so it is obvious he has to be willing to cooperate. b) If the catering manager would like to anticipate on the number of guests that will arrive, he should know early on how many people will arrive with regard to the quantity of food he has to prepare in advance, i.e. 09:00 – 09:30 in the morning. People in the building do not know yet if they are going to have lunch there or not. For that reason, an assumption is made that a reasonable stable percentage (with a limited variation) of the people in that building will have lunch. Based on that assumption, it is sufficient to know how many people are present in the building, which is not so difficult. In other cases, that fact is not known, available or disclosed to the caterer. c) When the amount of people in the building is known, the catering manager should make use of a simple tool (which is explained later on) to calculate how many food he has to prepare. This results in a certain weight for a specific product, for instance salad bar 4,500 grammes. The catering manager has to calculate how to divide this weight between the combinations of the salads that are offered. In view of the advantage this will provide, this will be used for the pilot, despite of the complexity. This has been performed in a limited form, namely one location per caterer and for one product, during a period of 3 weeks.

Before presenting the outcome, the simple tool is explained that catering managers could use: Day | Amount of people in the building | Product | Total amount offered | Loss | Eaten and offered – loss | Eaten per PERSON IN THE BUILDING. Not per lunch user (gram) | Offered per PERSON IN THE BUILDING, not per person using lunch (gram) | Monday | 200 | Part-baked rolls | 1,620 | 375 | 1,245 | 6.2 | 8.1 | Tuesday | 234 | Part-baked rolls | 1,830 | 425 | 1,405 | 6.0 | 7.8 | Wednesday | 168 | Part-baked rolls | 1,630 | 570 | 1,060 | 6.3 | 9.7 | Thursday | 261 | Part-baked rolls | 2,130 | 468 | 1,662 | 6.4 | 8.2 | Friday | 151 | Part-baked rolls | 1,775 | 730 | 1,045 | 6.9 | 11.8 |

Day | Number of people in the building | Product | Total amount offered | Quantity offered in grammes | Monday | 214 | Part-baked rolls | 1,481 | 6.9 | Tuesday | 245 | Part-baked rolls | 1,696 | | Wednesday | 140 | Part-baked rolls | 969 | | Thursday | 230 | Part-baked rolls | 1,592 | | Friday | 100 | Part-baked rolls | 692 | |
Table 14: tool for calculating the quantity of food that has to be prepared

Suppose a catering manager has a lot of waste after part-baked rolls, then that would be the selection in the pilot. The grey part of the table is filled in by the catering manager in week 1, in which he will work as he is used to. He will be informed about the number of people in the building. He will list how many part-baked rolls he will offer in total, i.e. all rolls added up in weight including the ones he will finish off in the oven at a later stage in time. The tool will automatically calculate how many grammes has been consumed per person in the building (so not per guest at lunch!) The maximum of that week (in this case 6.9 grammes) will be the basis for week 2 and 3. If 214 persons have lunch in week 2, then it is advised to prepare 214 x 6.9 = 1,481 grammes of part-baked rolls. The same procedure for the next 9 days in the pilot. As the grey part is also filled in for week 2 and 3, the effect on food waste is easily measured.

The introduction of the pilot without assistance, similar to the first phase of the project, proved to be very difficult. Only one result could be used and that is the reason why a new attempt has been made with yet another usable case. The complexity was not the only problem, as other reasons have also made the result poor during this test.

The two cases with regard to products that contribute most to waste are the following: Case 1: the salad bar with dressing: | Day | Amount in building | Offer | Loss | Loss % | Offer per inhabitant | Loss per inhabitant | Difference | Week 1 | Mon | 221 | 11,209 | 2,281 | 20.3% | 50.7 | 10.3 | 40.4 | | Tue | 218 | 11,385 | 1,076 | 9.5% | 52.2 | 4.9 | 47.3 | | Wed | 220 | 8,219 | 1,100 | 13,4% | 37.4 | 5.0 | 32.4 | | Thu | 234 | 11,067 | 1,300 | 11.7% | 47.3 | 5.6 | 41.7 | | Fri | 198 | 10,870 | 2,828 | 26.2% | 54.4 | 14.3 | 40.2 | Week 2 | Mon | 224 | 10,598 | 650 | 6.1% | 47.3 | 2.9 | 44.4 | | Tue | 217 | 7,304 | 0 | 0.0% | 33.7 | 0.0 | 33.7 | | Wed | 202 | 8,115 | 150 | 1.8% | 40.2 | 0.7 | 39.4 | | Thu | 231 | 2,500 | 150 | 6.0% | 10.8 | 0.6 | | | Fri | 194 | 7,337 | 200 | 2.7% | 37.8 | 1.0 | 36.8 | Week 3 | Mon | 226 | 9,850 | 910 | 9.2% | 43.6 | 4.0 | 39.6 | | Tue | 228 | 10,850 | 1,462 | 13.5% | 47.6 | 6.4 | 41.2 | | Wed | 214 | 9,000 | 600 | 6.7% | 42.1 | 2.8 | 39.3 | | Thu | 231 | 13,400 | 1,270 | 9.5% | 58.0 | 5.5 | 52.5 | | Fri | 197 | 12,946 | 4,197 | 32.4% | 65.7 | 21.3 | 44.4 |
Table 15: pilot results of the improvement measure for the salad bar with dressing

Note: as the yellow cell contains a rather deviating figure, this has not been included.
Analysis: in week 1, during which people worked as they used to, the offered products per resident in the building was high. In week 2, the products offered was lowered drastically, less than indicated in the tool. The effect was that loss almost disappeared that week. It is clearly visible that the consumption per person present (we refer to the column “difference”) decreased that week. This indicates that not enough products were offered and this resulted in revenue loss. It is logical that if you offer nothing, the waste is less as well, but this should not take place at the expense of turnover and/or profit. In week 3, this was taken into account except on the last day. The loss increased a bit there, but still less than in week 1. Visual results have been depicted in Figure 9.

Figure 9: waste in grammes per person present in the building during 3 weeks
Loss per inhabitant (gr)

Case 2: large pan of soup

| Day | Amount in building | Offer | Loss | Loss % | Offer per inhabitant | Loss per inhabitant | Difference | Week 1 | Mon | 448 | 48,000 | 3,700 | 7.7% | 107.1 | 8.3 | 98.9 | | Tue | 434 | 48,500 | 3,200 | 6.6% | 111.8 | 7.4 | 104.4 | | Wed | 484 | 40,000 | 3,800 | 9.5% | 82.6 | 7.9 | 74.8 | | Thu | 382 | 48,000 | 2,800 | 5.8% | 125.7 | 7.3 | 118.3 | | Fri | 286 | 40,000 | 3,300 | 8.3% | 139.9 | 11.5 | 128.3 | Week 2 | Mon | 386 | 44,000 | 2,600 | 5.9% | 114.0 | 6.7 | 107.3 | | Tue | 408 | 48,000 | 3,600 | 7.5% | 117.6 | 8.8 | 108.8 | | Wed | 332 | 42,000 | 1,900 | 4.5% | 126.5 | 5.7 | 120.8 | | Thu | 399 | 46,000 | 2,250 | 4.9% | 15.3 | 5.6 | 109.6 | | Fri | 287 | 38,000 | 2,000 | 5.3% | 132.4 | 7.0 | 125.4 |
Table 16: pilot results of the improvement measure for soup

Note: this pilot has been filled in for a period of 2 weeks

Analysis: in week 1, the loss is slightly more in comparison with week 2. This catering manager also did not use the tool, but carefully anticipated on the number of people in the building. As he has not reached the (safe) limit, the loss has only been limited from 7.36% average in week 1 to 6.02% for week 2.

Figure 10: waste in % of the offer during 2 weeks
Loss % of products offered

Conclusion of the pilots: it is clear that this improvement measure requires a lot of assistance during the execution. It is our experience with caterers that every change that is implemented on a large scale, is a difficult task which requires a lot of attention and energy. It seems a paradox but for that reason, the implementation of the tested improvement measure is a good proposal. After all, many improvement measures are product-related (for instance the entrance quality of fruit should be improved). This measure could lead to more opportunities for the reduction of food waste as the information is relevant for all products and basically, only one effort is necessary to achieve this.
An alternative to be able to estimate the number of guests at lunch is keeping track of the historical information based on the number of transactions on the cash register. The effect of this method depends on the dispersion of the information and will probably work on some locations and not work on others.
Furthermore, is it quite simple to understand that working based on historical data is not anticipating on the reality of the day so that sudden absence of groups of people on a day will always produce a lot of waste. Advantage of this method of working is the fact that the employer does not have to be burdened.

5 Conclusions and recommendations

An imperative condition to address the food waste problem, is knowledge about the exact location and cause of waste. In order to do this properly, information about the product category is not sufficient. It is necessary to gain insight about the quantity and causes per product.
The following table indicates the waste of the products that are contributing most. | Category | Product | Kg/wk | % of total | Cum% | CO2 (kg)/kg | CO2 (kg)/wk | 1 | Soup | large pan self-service | 34,446 | 33.2% | 33.2% | 3 | 10,3339 | 2 | Potatoes, vegetables and fruit | salad bar with dressing | 12,540 | 12.1% | 45.3% | 1.2 | 15,048 | 3 | Dairy | (butter)milk in jar | 8,021 | 7.7% | 53.0% | 1.6 | 12,833 | 4 | Potatoes, vegetables and fruit | salad bar without dressing | 7,858 | 7.6% | 60.6% | 0.5 | 3,929 | 5 | Bread | separate slices, soft white rolls | 6,394 | 6.2% | 66.7% | 0.7 | 4,476 | 6 | Bread | luxury rolls (croissants, hard crusted rolls) baked yourself | 3,985 | 3.8% | 70.6% | 0.7 | 2,789 | 7 | Bread | filled sandwiches | 3,092 | 3.0% | 73.6% | 0.7 | 2,164 | 8 | Meat | meat snack | 2,776 | 2.7% | 76.2% | 9.9 | 27,486 | 9 | Dairy | Boiled eggs | 2,452 | 2.4% | 78.6% | 1.4 | 3,433 | 10 | Potatoes, vegetables and fruit, 2/3 and meat 1/3 | on sandwich | 1,987 | 1.9% | 80.5% | | |

In total, about 104 tonnes per week is wasted in the catering sector. Soup from a large pan is largest in weight and also largest with regard to the environmental impact. The products in the category fish play an insignificant role, also because fish is not offered so much.

The most important cause of waste is the fact that it is unknown how many guests will have lunch. This cause is relevant for all products and that is the reason why this was tested in the pilot. The implementation of this measure has proven to be complex, as a result of which the outcome of the pilot has only an indicative value. This outcome suggests that anticipation on the number of people in the building will reduce the amount of food waste.

It is advisable to extend the outcome of this study further in the sector. In particular because the measuring methods are available now and experience is gained. It is wise to repeat this every year, for instance for the products soup and salad bar with dressing and possibly involve other products that contribute substantially to waste at a later stage. It is possible that Veneca might play a role here for standardization, so that on a sector level, the outcome can be monitored anonymously. Obviously, it is important that caterers start using the established causes and act upon this so that measuring will make the effects visible.

Recommendations to reduce food waste are: * A practicable method to make a better estimation of the number of guests, if need be based on historical data (pass on special occasions at the least) * Return the forced variation from the contract on the menu cycle and adapt it to the local situation based on consumers preferences. * Adding the required amount of products in doses towards the end of the lunch * Prepare after the order (bringing the product to the table later on (feeling of service) or a short waiting period * Adapt minimal order quantities for products which cause waste (butter tubs) * Processing food that is left over in other products (the madiwodo sandwich) * Tackling the incentive catering manager who is wrongfully purchasing products with a large margin in open book contracts

6 Acknowledgement

During the first phase of the project, many interviews were held and a lot of data has been measured. A number of people really contributed to this process, for which I would like to express my gratitude. I would like to thank the following people most kindly: Mariska Nijenhuis, Nancy Holthuysen, Nina Wassenaar, Manon Mensink en Milou Vrijhof, all working at Wageningen UR Food & Biobased Research

Literature

[ 1 ] FSIN, Foodservice Monitor Annual Report, 2010
[ 2 ] Meal service Máx à la Carte , 2011, http://www.totalcateringmoerdijk.nl/downloads/rapport-maxalacarte.pdf [ 3 ] Reduction of food waste in the Rijnland hospital, 2012
[ 4 ] www.veneca.nl

Enclosure A Interview table product category

A fresh fruit
B on sandwich
C salad bar with dressing
D salad bar without dressing
E Hot meal
F in dessert
G in bowl with dressing
H in bowl without dressing
I cold snack
J other, i.e. soup…

Way in which product is offered
Fill in “x” if product is offered this way
ON HOW MANY DAYS OF THE WEEK THE PRODUCT IS OFFERED THIS WAY
Top 3 with regard to loss in weight
Way of delivery with potatoes, vegetables, fruit (several answers are possible)
A fresh, processed (e.g. cut, in mix)
B fresh, unprocessed
C deepfreeze
D processed, ready-to-use on consumer level
E Dry groceries in a can
2 packaging size
Large-scale packing also containing
A consumer quantity
B large-scale packing, containing no further packing
Processing on several locations (several answers are possible)
A washing
B cutting
C making portions
D cooking

Fill in the below mentioned questions is “x” is filled in on the above-mentioned questions (DO NOT FILL IN TEXT BELOW, for example, after “other”: only fill in “x”)

4. At which moments, it is decided to add products?
a. not
b. if the products are used up
c. if half of the products are used up
d. if 1/3 of the products is left
e. other
5. What happens to the products that remain?
a. are thrown away
b. are processed in another product
c. are offered in the same way the next day
d. personnel uses these products for their lunch or take them home
e. other
6 If you have filled in 5a or 5d, how is this registered? a. Not b. Per consumer portion on cash register as money c. Per consumer portion, registered in a different way, …
7. If you have filled in 5a, how is it disposed of?
a. in the organic waste container / trash can
b. with the rest of the waste
c. other destination outside of the catering location

Enclosure B Example of filled in table of a catering location

Category bread

Way in which the product is offered | Top 3 | Day 1 | Day 2 | Day 3 | Day 4 | Day 5 | Day 6 | Day 7 | Day 8 | Day 9 | Day 10 | Date | | | 20-mei | 23-mei | 24-mei | 25-mei | 26-mei | 27-mei | 30-mei | 31-mei | 1-jun | Separate slices, soft white rolls, currant bun | n | 650 | 800 | 380 | 450 | 750 | 300 | 600 | 550 | 800 | 750 | Filled sandwiches | □ | 350 | 600 | 250 | - | - | 320 | - | 430 | - | 450 | Large loaf of bread (baked by baker) | □ | 260 | 150 | 270 | 480 | 210 | 430 | 320 | - | - | 440 | Large loaf of bread (baked yourself) | n | 870 | 250 | 400 | 700 | 350 | 450 | 750 | 900 | 320 | 300 | Luxury bread (baked by baker) | □ | 430 | 230 | 150 | 420 | 110 | 360 | 223 | 424 | 500 | 220 | Luxury bread (baked yourself) | □ | 400 | 150 | 220 | 150 | 130 | 260 | 210 | 150 | 250 | 405 | Other | □ | - | - | - | - | - | - | - | - | - | - |

13 separate slices + soft white rolls
1 large bread (baked by baker)
6 luxury bread (baker)

Enclosure C Results workshop with regard to improvement measures to reduce food waste | | Keeping score: estimated chance of success | A | Better insight into the number of visitors | high | medium | low | | Knowing how many people use lunch in advance | 1 | | | | Customers should inform us about special occasions (prior to incentive) | 5 | 3 | | | Keeping track of amount of visitors on certain points in time (via security) | | 2 | | | Order by a consumer in advance | | 1 | | | TOTAL | 6 | 6 | 0 | B | Range of products adaptation | | | | | Small portions | 1 | | | | Core range of products | | 1 | | | Smaller range of products | 1 | 2 | | | More madiwodo-soup/sandwich | 3 | | | | Offer more packed products | 1 | | | | No fruit anymore | 2 | | | | Eating what is cooked | 2 | 1 | | | Sell smaller portions | | 1 | | | Adapt range of products to the weather | 1 | 1 | 1 | | TOTAL | 11 | 6 | 1 | C | Adapt method of working on location | | | | | Start offering products at a later point in time | 1 | | | | Add products during lunch | 4 | | | | Prepare food during lunch | 2 | | | | Gone = gone | 3 | | | | Making soup at a later point in time | 1 | | | | No menu cycle anymore | 1 | 1 | | | A la minute, front cooking | | 2 | | | Production and sale at the same moment | | 1 | | | Snacks and filled sandwiches only after order | 1 | 1 | | | Daring to give no for an answer | 1 | | | | Smaller containers towards the end | 3 | 1 | | | Taking portions out of the freezer at different points in time | 1 | | | | Easier access to products with a lot of loss via routing | 1 | | | | Products on “use before” date in processed products | | 1 | | | TOTAL | 19 | 7 | 0 | D | Issues having to do with the distributor | | | | | Order unit | | 1 | | | Daily deliveries | | 1 | | | Controlling the quality of products (fruit) | | 1 | | | Adapting the minimum order units (smaller) | | 1 | 1 | | TOTAL | | 4 | 1 | E | Consumer | | | | | Drawing attention to loss with guests | 1 | 1 | 1 | | Deduct lunch money on salary automatically | | 1 | 1 | | Consumer investigation | | 1 | 1 | | Attracting consumers with product | 1 | | | | Keeping score of the loss per week (visualisation on location for everyone) | 2 | 1 | | | Show consumer the amount of waste | | 1 | | | Waste disposal fee for what is thrown away per consumer | | 1 | | | Knowing the preference of your consumer | | 2 | 1 | | Barometer on location | | 1 | | | Let consumers know you take action to avoid waste | 1 | | | | Walk for one hour | | | 1 | | TOTAL | 5 | 9 | 5 | F | Matters for the customer | | | | | Adapt the programme with requirements | 1 | 4 | | | More open book contracts | | 1 | | | Agreements with customers about range of products on time | 1 | 1 | | | Lower prices at the end of the lunch | 2 | 1 | | | Let customer communicate policy about waste | 1 | | | | TOTAL | 5 | 7 | 0 | G | Matters concerning catering personnel | | | | | Education of own employees | | 3 | 2 | | Reward catering manager for loss reduction | 1 | | 1 | | Creating awareness in own catering personnel | 1 | 2 | | | TOTAA | 2 | 5 | 3 | H | Process-related/technical | | | | | Orange squeezer | 1 | | | | Develop machine for smoothies (using leftover products) self-service | 1 | | 2 | | Suggestion box for leftover products | | 1 | 1 | | Processing leftover products | 2 | | | | Integration of leftover products in menu planning | 3 | | | | Equipment that avoids waste: bread-slicers, defrosting machines etc. | | | 1 | | TOTAL | 8 | 1 | 4 | I | Various | | | | | Insight on parameters on which no influence is possible on a location level | | 1 | | | Feedback of information about sales results | | 1 | | | Showing no actual products but pictures of products | | 1 | | | TOTAL | 0 | 3 | 2 |

Enclosure D Food waste survey in catering sector | Category | Product | Kg/wk | % of total | Cum % | CO2 (kg)/kg | CO2 (kg)/wk | 1 | Soup | Large pan, self-service | 34,446 | 33.2 | 33.2 | 3 | 103,339 | 2 | Potatoes, vegetables, fruit | Salad bar with dressing | 12,540 | 12.1 | 45.3 | 1.2 | 15,048 | 3 | Dairy | (Butter)milk in jar | 8,021 | 7.7 | 53.0 | 1.6 | 12.833 | 4 | Potatoes, vegetables, fruit | Salad bar without dressing | 7,858 | 7.6 | 60.6 | 0.5 | 3,929 | 5 | Bread | Separate slices, soft rolls | 6,394 | 6.2 | 66.7 | 0.7 | 4,476 | 6 | Bread | Luxury rolls, croissants, hard crusted rolls, self-baked | 3,985 | 3.8 | 70.6 | 0.7 | 2,789 | 7 | Bread | Filled sandwiches | 3,092 | 3.0 | 73.6 | 0.7 | 2,164 | 8 | Meat | Meat snack | 2,776 | 2.7 | 76.2 | 9.9 | 27,486 | 9 | Dairy | Boiled eggs | 2,452 | 2.4 | 78.6 | 1.4 | 3,433 | 10 | PVF (2/3) and meat (1/3) | On sandwich | 1,987 | 1.9 | 80.5 | | | 11 | Dairy | Added value dairy drink | 1,931 | 1.9 | 82.4 | 1.6 | 3,090 | 12 | Meat | Meat unpacked, separate | 1,779 | 1.7 | 84.1 | 18.8 | 33,450 | 13 | Soup | Soup in bowls | 1,758 | 1.7 | 85.8 | 3 | 5,275 | 14 | (3/4PVF, ¼ meat) | Hot meal | 1,680 | 1.6 | 87.4 | | | 15 | (1/2PVF,1/4 bread, 1/4dairy) | Other | 1,170 | 1.1 | 88.5 | | | 16 | Dairy | (Butter)milk in dispenser | 1,146 | 1.1 | 89.6 | 1.6 | 1,834 | 17 | PVF | Fruit salad | 1,056 | 1.0 | 90.6 | 0.3 | 317 | 18 | Dairy | Margarine and butter tubs | 969 | 0.9 | 91.6 | | | 19 | Dairy | (Butter) milk prepacked | 948 | 0.9 | 92.5 | 1.6 | 1,517 | 20 | Dairy | Prepacked dessert | 914 | 0.9 | 93.4 | 1.1 | 1,005 | 21 | PVF | Fresh fruit | 868 | 0.8 | 94.2 | 0.5 | 434 | 22 | Meat | Meat, packed | 843 | 0.8 | 95.0 | 18.8 | 15,846 | 23 | Bread | Luxury rolls (croissants, hard-crusted rolls) baked by baker | 833 | 0.8 | 95.8 | 0.7 | 583 | 24 | Bread | Meat loaf, self-baked | 753 | 0.7 | 96.6 | 0.7 | 527 | 25 | Dairy | Dessert, self-made | 676 | 0.7 | 97.2 | 0.9 | 608 | 26 | Meat | Added in soup yourself | 567 | 0.5 | 97.7 | 4.3 | 2,437 | 27 | PVF | In a bowl without dressing | 392 | 0.4 | 98.1 | 0.5 | 196 | 28 | Bread | Loaf of bread, baked by baker | 389 | 0.4 | 98.5 | 0.7 | 273 | 29 | Dairy | Cheese, separate | 356 | 0.3 | 98.8 | 11.5 | 4,097 | 30 | Fish | Fried fish (cod parings) | 237 | 0.2 | 99.1 | 2 | 475 | 31 | Fish | Fish on bread (salmon) | 221 | 0.2 | 99.3 | 2 | 441 | 32 | Dairy | Snack with dairy | 176 | 0.2 | 99.5 | 12 | 2,108 | 33 | Dairy | Cheese on a roll | 158 | 0.2 | 99.6 | 11.5 | 1,814 | 34 | Dairy | Fried eggs | 119 | 0.1 | 99.7 | 1.3 | 154 | 35 | PVF | Cold snack | 109 | 0.1 | 99.8 | 1.2 | 131 | 36 | Meat | Added in a salad yourself | 89 | 0.1 | 99.9 | | | 37 | Fish | Cold fish | 64 | 0.1 | 100 | 2 | 128 | 38 | Dairy | Cheese prepacked | 26 | 0.0 | 100 | 11.5 | 295 | 39 | Fish | Other (fish snacks) | 0 | 0.0 | 100 | | | 40 | PVF | In a bowl with dressing | 0 | 0.0 | 100 | 1.2 | 0 | | | TOTAL | 103,778 | 100% | | | 252,532 |

--------------------------------------------
[ 1 ]. Vereniging Nederlandse Cateringorganisaties (Association of Dutch catering organisations), www.veneca.nl
[ 2 ]. http://www.levensmiddelenkrant.nl/2137/CBL:%20'Als%20eerste%20food waste%20aanpakken'
[ 3 ]. WUR is currently executing this assignment
[ 4 ]. Veneca = vereniging Nederlandse cateringorganisaties (Association of Dutch Catering Organisations), www.veneca.nl
[ 5 ]. Other investigations have been performed by WUR/FBR in hospitals where the amount of waste in traditional kitchens is relatively much higher than mentioned in this project (we refer to [2] and [3])
[ 6 ]. Both location type ‘school’ and meal type ‘lunch+dinner’ was not present in combination with < 100 lunch users
[ 7 ]. At the time we performed the measurements, this was not part of Compass yet
[ 8 ]. Please notice that product level does not correspond with the SKU from retail in this report. The fact of the matter is that it is not unique. A filled sandwich has been labelled product level in this project, but many types of filling naturally exist.
[ 9 ]. In another project in the catering … it was investigated what the effects of CO2 labelling are on purchasing behaviour. For that purpose, CLM has made a global, integral analysis of a number of products in the catering up to and including offering products on location. For a number of our 45 products, this revealed the CO2 equivalent.
[ 10 ]. Some experts indicate a maximum of 2 to 3 minutes while others claim you could also bring the product later.
[ 11 ]. The data to investigate this further is available but selections have been made what to investigate further and what not. After all, the database offers sufficient space for many cross-sections in many different areas. The most relevant matters for the project group are included in this report.
[ 12 ]. In this case, the catering manager has acted upon impulse and offered less without using the tool as guiding principle.