Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Food science

Comparison of relative energy costs of village-prepared and centrally processed weaning foods
High-protein biscuits made with ragi flour and oil-seed flour blends
A Simplified procedure for calculating amino-acid scores of blended foods or dietary patterns
A simple field kit for testing iodine in salt


Comparison of relative energy costs of village-prepared and centrally processed weaning foods

Judson M. Harper and Ronald E. Tribelhorn
Colorado State University, Fort Collins, Colorado, USA


One cause of weaning-age malnutrition in developing countries is the complementation of breast milk with cereal gruels that are not calorie-dense. These gruels usually have poor nutritional qualities and lack many benefits provided by a weaning food specifically designed for the infant's nutritional needs [6]. There is no literature available that compares the energy required to prepare centrally processed weaning foods in the home with that required to prepare conventional weaning foods.

The purpose of this article is to compare the in-home energy requirements and costs of preparing two types of weaning foods, both made from the same ingredients, but processed differently before being used. The article also examines energy differences between single servings and multiple servings using the two preparation methods. The preparation methods for multiple servings would be similar to those used in schools or health clinics.

Table 1 lists the more obvious advantages and disadvantages of the two preparation approaches. Perhaps the most significant advantage of pre-cooked foods is their convenience for the mother. Since conventional methods involve the preparation of foods from raw grains, the homemaker may be reluctant or unable to make a special weaning food as this requires considerable time and attention.

The enhanced nutritional value of added vitamins and minerals as well as the increased caloric density of centrally processed foods are additional advantages [6]. Typically, foods prepared in the home using conventional methods do not have these characteristics because of the unavailability of ingredients and their high cost.

The advantages of conventionally prepared weaning foods include the use of a variety of raw materials and the fact that the food is boiled as part of the cooking, thus reducing potential contamination. The instant preparation of centrally processed food can involve the use of unboiled tap water, often unsuitable because it may contaminate the prepared product.

TABLE 1. Advantages and disadvantages of preparation techniques

Centrally processed
Conventionally prepared
1. Convenient to use 1. Lower cost
2. Higher calorie density 2. Minimal packaging
3. Fortified with vitamins
and minerals
3. No additional distribution
4. Long shelf life since
product is dry
4. Variation in ingredients
5. Cooking required reduces
risk of contamination
5. Image
6. Tradition and familiarity  
1. Separate packaging 1. Preparation time
2. Cost may be somewhat
2. Lack of utensils to cook
3. Unsanitary if prepared
with non-potable water
3. May be only partially
4. Distribution required 4. Short shelf life following
5. Long-term availability
6. Price stability 5. No opportunity to add
vitamins and minerals
  6. Fuel-intensive


Figure 1 outlines the two weaning food systems examined in this study. Appropriate assumptions about each system's energy inputs have been made to simplify the analysis. Both systems have in common: grain cleaning, dehulling, grinding, packaging, and transport. It was assumed that these common steps in each system require the same amounts of energy per unit of product. Packaging was included since both raw or processed materials might be purchased in similar packaging material.

FIG. 1. Two alternate processes for producing weaning foods

The home-preparation procedures are different for the systems studied, making this step the key to the analysis. In this study, the comparison is reduced to an examination of the preparation energies needed for conventional homemade foods versus those needed for the grinding, cooking, and cooling steps in centrally processed food production as well as for preparing the product in the home.

In judging the validity of this study, the following additional assumptions are important:

1. The electrical energy used in both systems was from a common source.
2. Secondary or more basic energy levels, such as energy expended in the manufacture of equipment and in the extraction and production of energy resources, and energy associated with agricultural production, were not considered.
3. It was assumed that, prior to the purchase of raw materials, the same amounts of energy were required to produce, transport, etc., all raw materials used in both systems.
4. Packaging materials may have been different for the two systems studied. However, we assumed packaging materials and unit quantities were the same in each system.
5. Each home preparation system involved restarting the heat source used.
6. The raw materials were purchased pre-processed (cleaned, dehulled, and milled} from suppliers.
7. Centrally processed foods were not fortified with vitamins, minerals, and milk powder so as to make them more comparable to the conventionally prepared products.
8. The study was limited to low-cost extrusion [5] for the central processing of weaning foods.
9. Home cooking utensils were used correctly so efficient heating occurred when different heat sources were applied. 10. The ecological effects of the alternative methods were not considered. For example, these might include depletion of wood supplies and subsequent soil erosion associated with wood-burning cooking practices.

Literature on energy accounting frequently points out that, in energy comparison analyses, failure to examine secondary or more basic levels of energy may lead to inaccurate calculations. Typically, an incorrect assumption can result in a significant energy-intensive step being omitted from the analysis, which affects the final energy accounting for each system. Should this prove to be the case for this study, a more thorough investigation may be required to quantify accurately the actual energy inputs of the alternative systems for preparing weaning foods, but such an analysis was considered beyond the scope of the present work.


Information on the energy used by the central processing system was available from the records of existing plants and the results of previous investigations [4, 5, 8] . Information on the energy requirements for the conventional preparation of weaning foods was limited; therefore, as a model we used a simple cereal/legume combination recipe [3] that specified that material preparation and cooking procecures were to be carried out in the home (Appendix 1). The recipe included not only the single-serving size but also multiple-serving quantities, making it possible to study large serving sizes such as might exist at the village level or in large feeding programmes. For the purposes of the study, soya beans were substituted for the dry beans specified in the recipe to increase the similarity in caloric densities between the conventionally prepared food and the centrally processed food, which typically uses soya bean. The cooking time for the conventional preparation was adjusted to eliminate raw flavours in the mixture. Water was added as required (subjectively) to adjust viscosity so that the cooked gruel would have a consistency suitable for spoon-feeding to a weaning-age child.

Several energy sources are used for cooking by lower-income people in developing countries. Most often wood or kerosene is used. Quantifying the use of wood as a heat source is difficult since the procedures used for building a fire and cooking differ. In addition, the efficiency of heat utilization is greatly dependent on the procedures followed.

More accurate and consistent data are possible when using energy sources that can be controlled and accurately measured, as is the case with both electricity and gas. Although wood was used as a fuel source in this study, more attention will be given to electricity and gas for comparison purposes because their usage can be accurately quantified.

The procedures used to prepare foods are discussed for each energy source below.


The following equipment and materials were used to measure energy for preparation in both procedures.
1. Aluminium saucepan (1 litre, uncovered).
2. Aluminium saucepan (3 litres, uncovered).
3. Electric range heating coil selected to match diameter of pan.
4. Stop watch.
5. Watt hour meter.
6. Raw materials-rice, soya beans, sugar.

The recipe given in Appendix I was followed to prepare the food. The following procedure was used to determine cooking energies.
1. Connect watt-hour meter in series with stove.
2. Weigh ingredients for food preparation.
3. Add water to pan and heat over coil (on high) to boiling.
4. Add remaining ingredients. Bring to boil.
5. Reduce heat to simmer.
6. Cook until raw flavour is gone.
7. When cooking is complete, turn off heating coil and
record data.
8. All preparations were started with a cold heating coil.

It was necessary to add water during the simmering stage to maintain a reasonable viscosity in the product. Quantities of water were recorded.


The equipment and materials used for this analysis were similar to those listed above, with the following exceptions:
- The electric range was replaced with a propane burning stove.
- A balance was used to measure the propane fuel consumed.

The preparation procedure followed was the same as above with the weight of the propane consumed and the amount of water added being recorded.


A two-litre cast-iron Dutch oven was suspended over an open pit fire so that a majority of the flame points were in contact with the pan surface. The fire pit was approximately 60 cm square and 20 cm deep. Enough wood was weighed to start a fire which seemed suitable for the size of pot. The wood weight and moisture were recorded prior to starting the fire. Wood was added during the preparation to maintain a uniformly sized flame throughout the test. The total weight of the wood consumed was recorded during the cooking process along with the amount of water added to maintain consistency.

The quantity of wood needed to start the fire was recorded; however, the wood used was not assigned to any particular preparation size since wood cooking fires usually burn throughout the day.


The energy consumption and the specific energies (heat energy per gram of food solid) for each heating method was determined. Energy consumption was converted to monetary values using the following cost factors:
1. Electric: US$1.7 x 10-5/kJ (US$0.06/kWhr)
2. Gas: US$7.5 x 10- 6 /kJ (US$0. 1 93/kg)
3. Wood: US$2.88 x 10-6/kJ (US$55/cord)


Serving Size

The quantity of weaning food prepared was found to have a direct effect on the cooking energy requirement. Using package directions for preparing a single portion of the centrally processed food resulted in a quantity approximately one-half the size of that made following the conventional recipe in Appendix I. Under these circumstances, the specific energy requirements for heating the smaller quantity of centrally processed gruel were found to be higher than those needed for the preparation of a larger portion. These higher specific energy requirements were due to the efficiency with which the pan and contained mass captured the heat from the source. This result seems to be in agreement with other research findings [7] where the energy to cook any dry flour product was found to be directly related to the quantity of water required.

Figure 2 shows the quantity of electrical energy required to boil a specified volume of water using a standard 1.1-litre container. Using the same cooking vessel, approximately one-half as much energy per kilogram was required to boil 275 ml of water as to boil 138 ml of water. Similar trends resulted when gas was used as the heat source. It may be concluded that the size of preparation has a direct effect on the energy required and necessarily on the cost.

Energy Requirements

Energy usage measured during the tests and normalized to the weight of the solids in the mixture is given in table 2. These data are given for preparation sizes of one, two, three, five and ten servings with the serving size set at 280 grams of prepared food. Using FAO tables [2], 280 grams would contain approximately 15 grams of protein and375 calories for a rice soya bean mixture such as the one used in this investigation.

FIG. 2. Energy effects when changing prepared quantity of centrally processed gruel

TABLE 2. Energy and cost comparison of alternate gruel preparationsa

  Energy components kJ fuel energy/gram solids    
Preparation time (mins.) Energy cost per dayb,c
Number of servings Serving size Central Home Total
A. Single serving  
1. In home, electric 0.28 0 17.8 17.8 25.0 0.0048
2. In home, gas 0.28 0 25.1 25.1 26.0 0.0030
3. In home, wood 0.28 0 597.0 597.0 33.0 0.0017
4. Central Instant, electric 0.18 3.3 7.3 10.6 3.0 0.0029
5. Central Instant,gas 0.18 3.3 8.8 12.1 2.0 0.0015
6. Central Instant,wood 0.18 3.3 413.0 413.0 5.0 0.0012
7. Central Instant, electric 0.28 3.3 5.6 8.9 4.0 0.0024
8. Central Instant, gas 0.28 3.3 10.0 13.3 3.0 0.0016
B. Two servings  
1. In home, electric 0.56 0 13.0 13.0 27.0 0.0035
2. In home, gas 0.56 0 19.2 19.2 28.0 0.0023
3. Central Instant, electric 0.56 3.3 4.9 8.2 7.0 0.0022
4. Central Instant, gas 0.56 3.3 6.5 9.8 8.0 0.0012
C. Three servings  
1. In home, electric 0.84 0 11.3 11.3 27.0 0.0031
2. In home, gas 0.84 0 13.7 13.7 26.0 0.0016
3. Central Instant, electric 0.84 3.3 5.1 8.4 6.0 0.0023
4. Central Instant,gas 0.84 3.3 5.0 8.3 8.5 0.0010
D. Five servings  
1. In home, electric 1.40 0 10.0 10.0 31.0 0.0027
2. In home, gas 1.40 0 11.4 11.4 31.0 0.0014
3. Central Instant, electric 1.40 3.3 4.1 7.4 8.0 0.0020
4. Central Instant, gas 1.40 3.3 4.8 8.1 14.0 0.0010
E. Ten servings  
1. In home, electric 2.80 0 6.3 6.3 32.0 0.0017
2. In home, gas 2.80 0 9.6 9.6 32.0 0.0012
3. In home,wood 2.80 0 120.0 120.0 30.0 0.0003
4. Central Instant, electric 2.80 3.3 3.6 6.9 14.0 0.0019
5. Central Instant, gas 2.80 3.3 4.9 8.2 35.0 0.0010

a. Percentage solids; in home, 12.3 per cent; Central,16.5 per cent (16 grams solid = 270 kcal).

b. Based on 270 kcal per day average intake as supplement or 1/3total energy intake. Fuel costs: electricity, $1.7 x 10- s /kJ; gas, $7.5 x 10-6/kJ; wood, $2.83 x 10-6/kJ. Cost to feed child one year old or less.

c. Preparation energy values determined for wood as source of energy are related to building of the fire and stoking, which is a human judgement. Wood energy cost was valued according to current purchase costs of firewood. Realistic values cannot be assigned to wood because wood used by the home-maker typically comes from land nearby and is free.

Source: Cameron and Hofvander [1] .

The energy requirements for single-serving gruels prepared at home using conventional techniques from raw ingredients are shown under heading A (entries 1-3) in table 2. A 4-6 gives the energy requirements for the preparation of a centrally processed food using the quantities recommended on the package for a single feeding. A7-8 reflects an adjustment in the quantities of the centrally processed food to the same quantities used by the in-home food prepared from raw ingredients.

In preparing single servings there were significant differences between the methods studied in terms of the amount of energy required to make available one gram of food solids. The instant centrally processed formula required approximately one-half as much energy as the same amount of food made from raw materials. The reason for these differences is that the food made at home from raw materials required cooking for a long time with relative inefficiency in small containers.

Test results showed that the differences in energy usage between the two preparation techniques diminished as the serving size increased. The energy required to prepare the foods approached an asymptote, which was the lowest amount of energy required to cook the solids given the liquid-solid mixture for each particular energy source Details of these results are shown in figure 3

The high energy required to cook small serving size preparations is shown in figure 3 The relative energy differences between the two preparation methods indicate that centrally processed foods may offer an economic advantage for small preparation sizes. As the preparation quantity increases, there is less of an energy advantage for one method compared to the other.

Economic Considerations

Guidelines have been established concerning the amounts of food a child should consume [1] . These guidelines give the intake in kcal consumed per day independent of the food type. For example, the FAO recommends a calorie intake of 820 kcal/day for a child 6 to 12 months old [2] Assuming the child receives one-third of the total energy intake from weaning foods and two-thirds from breast milk, an economic analysis can be performed to determine the energy cost required for feeding a child on a daily basis

The results of such an economic analysis are shown in table 2; they indicate that the instant centrally processed food required approximately one-half the energy cost to cook a single serving compared to a similar preparation made from raw materials The overall cost saving (using electric energy as the energy source) was approximately one-fourth of one cent per day or $0.75 per month, which might be significant for some developing-country families.

FIG. 3. Energy requirements for different serving sizes

Lower energy requirements were measured for multipleserving preparations. Under these circumstances, the energy cost per gram of solids was reduced substantially, indicating a potential advantage in the use of co-operative units to prepare multiple food servings.

Similarities between Centrally Processed Weaning Foods and US Products

A close parallel between the alternative preparation techniques for weaning foods can be made with the cooking of rice and instant rice products in the United States. Instant rice is pre-cooked and only requires boiling water for its preparation. Raw rice must be boiled for several minutes in water to cook it completely. Thus, instant rice is similar to the centrally processed food while raw rice is similar to food that is conventionally prepared at home using raw materials. Because of the similarity, both raw rice and instant rice were cooked using electrical energy following the package directions. The results showed that raw rice required 3.7 kJ per gram of dry solids to cook, compared to 2.9 kJ per gram for instant rice. Using a gas energy source, raw rice required 7.4 kJ per gram of dry solids and instant rice 4.1 kJ. The relative energy differences between raw and instant rice compared favourably with the differences found between in-home preparation from raw ingredients and centrally processed weaning foods.


The results of this study showed that for single servings there were differences in energy requirements and costs between the preparation of a weaning food made with raw materials and that of a pre-cooked centrally processed weaning food. The centrally processed food required less energy/preparation per single serving. The differences diminished as the preparation size increased. On an absolut scale, preparation of large quantities of food always resulted in higher energy efficiencies. Depending on the type of food used and the serving size prepared, the energy cost of weaning food preparations may be significant in the developing country context.


Dried Beans-Rice Porridge

  For one For ten
Dried beans 2 tbsp. 11/2 cups
Rice 2 tbsp. 11/2 cups
Water 1 cup 8 cups
Milk powder (if available) 1 tbsp. 2/3 cup
Brown sugar 1 tbsp. 2/3 cup

1. Prepare dried bean flour and rice flour from well-dried beans and rice by pounding or grinding. If thorough drying in sunlight is not possible, heating gently over a low fire helps to yield a fine flour. (Flour of dried beans or rice may keep for one month in an airtight container, if the grains are first slowly toasted for 15 minutes until golden brown, then ground or pounded and quickly stored in the airtight container.) A flour which is more readily soluble in water can be made by boiling the beans for 30 minutes, then drying thoroughly and finally pounding or grinding.
2. Blend dried bean flour and rice flour in small amount of cold water.
3. Boil remaining water and add mixture, stirring constantly to prevent scorching.
4. Add sugar and milk, or fish flour (if available).
5. Cook for 15 minutes more to a soft custard consistency, adding more water if necessary.


1. M. Cameron and Y. Hofvander, Manual on Feeding Infants and Young Children. 2nd ed. (Protein-Calorie Advisory Group of United Nations System, United Nations, New York, 1976).

2. FAO, Handbook on Human Nutritional Requirements, FAO Nutritional Studies No. 28 (FAO. Rome, 1974).

3. P. R. Hale and B. D. Williams, eds., Litlik Buk (Wirui Press, Wewak, Papua New Guinea, 1977).

4. J. M. Harper and G. R. Jansen, eds., Low-cost Extrusion Cookers, First International Workshop Proceedings (Department of Agricultural and Chemical Engineering, Colorado State University, Fort Collins, Colo. 1976).

5. J. M. Harper, G. R. Jansen, J. D. Kellerby, and R. E. Tribelhorn, Evaluation of Low-Cost Extrusion Cookers for Use in LDC's, Annual Report, LEC-9 (Department of Agricultural and Chemical Engineering, Colorado State university, Port Collins, Colo., 1980).

6. G. R. Jansen, L. O'Deen, R. E. Tribelhorn, and J. M. Harper, "The Calorie Densities of Gruels Mode from Extruded Cornsoy Blends, " Food Nutr. Bull., 3(1): 39-43 (1981).

7. M. Rawitscher and J. Mayer, "Energy Requirements and Breakfast Cereals," Food Policy, 6: 27(1981).

8. D. E. Wilson and R. E, Tribelhorn, Low-cost Extrusion Cookers, Second International Workshop Proceedings (Department of Agricultural and Chemical Engineering, Colorado State University, Fort Collins, Colo., 1979).


Contents - Previous - Next