evidence-based blog of Filippo Dibari

Search for “linear programming”

The use of linear programming to determine whether a formulated complementary food product can ensure adequate nutrients for 6- to 11-month-old Cambodian infants

In Under-nutrition on February 5, 2014 at 11:46 am

Jutta KH Skau, Touch Bunthang, Chhoun Chamnan, Frank T Wieringa, Marjoleine A Dijkhuizen, Nanna Roos, and Elaine L Ferguson

Am J Clin Nutr January 2014 vol. 99 no. 1 130-138


Background: A new software tool, Optifood, developed by the WHO and based on linear programming (LP) analysis, has been developed to formulate food-based recommendations.

Objective: This study discusses the use of Optifood for predicting whether formulated complementary food (CF) products can ensure dietary adequacy for target populations in Cambodia.

Design: Dietary data were collected by 24-h recall in a cross-sectional survey of 6- to 11-mo-old infants (n = 78). LP model parameters were derived from these data, including a list of foods, median serving sizes, and dietary patterns. Five series of LP analyses were carried out to model the target population’s baseline diet and 4 formulated CF products [WinFood (WF), WinFood-Lite (WF-L), Corn-Soy-Blend Plus (CSB+), and Corn-Soy-Blend Plus Plus (CSB++)], which were added to the diet in portions of 33 g/d dry weight (DW) for infants aged 6–8 mo and 40 g/d DW for infants aged 9–11 mo. In each series of analyses, the nutritionally optimal diet and theoretical range, in diet nutrient contents, were determined.

Results: The LP analysis showed that baseline diets could not achieve the Recommended Nutrient Intake (RNI) for thiamin, riboflavin, niacin, folate, vitamin B-12, calcium, iron, and zinc (range: 14–91% of RNI in the optimal diets) and that none of the formulated CF products could cover the nutrient gaps for thiamin, niacin, iron, and folate (range: 22–86% of the RNI). Iron was the key limiting nutrient, for all modeled diets, achieving a maximum of only 48% of the RNI when CSB++ was included in the diet. Only WF and WF-L filled the nutrient gap for calcium. WF-L, CSB+, and CSB++ filled the nutrient gap for zinc (9- to 11-mo-olds).

Conclusions: The formulated CF products improved the nutrient adequacy of complementary feeding diets but could not entirely cover the nutrient gaps. These results emphasize the value of using LP to evaluate special CF products during the intervention planning phase.

– – –
NB – To follow up this topic (or others), enter your email in the rectangle at the bottom/right side of this page (you can un-subscribe any time).


In Under-nutrition on November 16, 2013 at 7:14 pm

by Andrew K. Amegovu, Patrick Ogwok, Sophie Ochola, Peter Yiga, Juliet H. Musalima, Emma Mutenyo

from Journal of Food Chemisty and Nutrition – Vol 1, No 2 (2013)



Infant and young child feeding practices in low-income countries are still inadequate leading to high rates of acute malnutrition. Formulas from local food materials are vital in formulations for management of child malnutrition in poor countries because they are affordable. Nutrient composition of sorghum-peanut blend (SPB) mixed with honey and ghee, and micronutrient-fortified corn-soy blend (CSB), a traditional food supplement, were analyzed. Proximate components and beta-carotene amounts were high in both products. Vitamin A level was higher in CSB than SPB. Proportions of essential fatty acids were low. Levels of iron, zinc, calcium, magnesium, phosphorus, potassium, manganese and sodium were adequate for recovery from moderate acute malnutrition (MAM). Energy content of CSB was 421kcal/100g while that of SPB was 430kcal/100g. Levels of condensed tannin, phytates, trypsin inhibitors and aflatoxins were below prescribed limits. In conclusion, levels of nutrients in SPB and CSB were adequate for treatment of MAM in children.

– – –
NB – To follow up this topic (or others), enter your email in the rectangle at the bottom/right side of this page (you can un-subscribe any time).

What linear programming contributes: world food programme experience with the “cost of the diet” tool.

In Under-nutrition on July 14, 2013 at 9:41 am

by Frega RLanfranco JGDe Greve SBernardini SGeniez PGrede NBloem Mde Pee S.

Food Nutr Bull. 2012 Sep;33(3 Suppl):S228-34.

Background – Linear programming has been used for analyzing children’s complementary feeding diets, for optimizing nutrient adequacy of dietary recommendations for a population, and for estimating the economic value of fortified foods.

Objective – To describe and apply a linear programming tool (“Cost of the Diet”) with data from Mozambique to determine what could be cost-effective fortification strategies.

Methods – Based on locally assessed average household dietary needs, seasonal market prices of available food products, and food composition data, the tool estimates the lowest-cost diet that meets almost all nutrient needs. The results were compared with expenditure data from Mozambique to establish the affordability of this diet by quintiles of the population.

Results – Three different applications were illustrated: identifying likely “limiting nutrients,” comparing cost effectiveness of different fortification interventions at the household level, and assessing economic access to nutritious foods. The analysis identified iron, vitamin B2, and pantothenic acid as “limiting nutrients.” Under the Mozambique conditions, vegetable oil was estimated as a more cost-efficient vehicle for vitamin A fortification than sugar; maize flour may also be an effective vehicle to provide other constraining micronutrients. Multiple micronutrient fortification of maize flour could reduce the cost of the “lowest-cost nutritious diet” by 18%, but even this diet can be afforded by only 20% of the Mozambican population.

Conclusions – Within the context of fortification, linear programming can be a useful tool for identifying likely nutrient inadequacies, for comparing fortification options in terms of cost effectiveness, and for illustrating the potential benefit of fortification for improving household access to a nutritious diet.

You can find more posts about linear programming clicking here.

– – –

NB – To follow up this topic (or others), enter your email in the rectangle at the bottom/right side of this page (you can un-subscribe any time).

Linear Programming in Nutrition: easy! Yes, but…only if somebody would have told me how to do it.

In Under-nutrition on May 3, 2012 at 11:40 am

This week I am working again for Valid International creating new formulations for feeding programmes in emergency settings. The software to create the theoretical formulation is a simple MS Excel sheet, using the add-in Solver. However the technique is based on linear programming : something very mysterious until few years ago for me, now much more accessible. Basically it is a mathematical approach to solve multi-factorial equations – scary isn’t it?

However, the potential use of Linear Programming (also) in nutrition is huge! You can design and assess diets and formulate new foods. Not only. As an example see this economic assessment of food prices using Linear Programming (Briend et al.). Or for the ones interested in therapeutic feeding, do not miss this paper.

For some simple theory  have a look at this UN document about  Linear Programming in nutrition, prepared by Andre’ Briend: very well done. If you are specifically interested on ready-to-use therapeutic food, see this recent paper. If you are looking for a simple software, try Nutrisurvey (the picture is from this software): a marvel work by Juergen Erhardt.

Question: are you a nutritionist or a food technologist? Did you receive a training in this amazing tool during your education? If not, why is it so? Do you think there is market out there for this?

My opinion – I had to learn myself… Interesting, but time demanding. As a nutritionist, or food technologist, my point is that this tool should be a compulsory part of our formal education (MSc, bachelor, etc.), just as much we learn about food and nutrient composition.

What is your point about it?



Find more information about LP also here.

Food allowance optmization model

In Under-nutrition on February 5, 2014 at 12:11 pm

by Viesturs Rozenbergs, Imants Skrupskis, Dace Skrupska, Ērika Rozenberga




Possibility of food allowance optimization by using MS Solver tool is analysed in the research. The model is developed by balancing 22 food products and 30 constraints – 8 nutrients and 22 minimum amounts of food products. The new method differs from the applications of linear programming described in the special literature on nutrition science not only with increased nutritional constraints, but also the minimum amount of every product is introduced as constraints, which does not essentially change costs, but provide quality, for example, for tea or coffee it is recommended to define not x≥0, but x≥3. By modifying minimum amounts of tea, coffee, sugar, spices, it is possible to obtain up to 70% economy from the initial rate. Application of the model is approbated in the computer class during practical classes for students of nutrition science.

– – –
NB – To follow up this topic (or others), enter your email in the rectangle at the bottom/right side of this page (you can un-subscribe any time).

How to design, pre-test and trial low cost, local RUTF products to rehabilitate severely malnourished children and adults

In Under-nutrition on August 9, 2012 at 3:55 pm

The latest number of Field Exchange, a journal published by ENN, contains a summary of a recent research paper describing how to use “Linear programming to design low cost, local RUTF” with the aid of Microsoft Excel software.

Whereas the original paper abstract can be found on The Journal of Nutrition web site, in the Field Exchange journal (issue 43, part 1), the article summary is available (for free) at page 36, including an “adapted” flow chart with the step sequence proposed for the design (see beneath), the pre-testing and the trialling of these products, currently in high demand in feeding programmes around the world.

The authors of the original paper are Filippo Dibari (author also of this blog), El Hadji I. Diop, Steven Collins, and Andrew Seal from Valid International and the University College of London.

More information about the Linear Programming, applied to the field of nutrition, is also available from another post in this same blog.

My publications

In on April 30, 2012 at 5:03 pm
(pdf files can be available under request – contact me)

NB – To follow up these list (or other topics of this blog) enter your email in the rectangle at the bottom/right side of this page (you can un-subscribe any time)

My PhD

In on April 21, 2012 at 8:10 pm

This slideshow requires JavaScript.

Title: “Design and pre-testing of lipid-based, ready-to-use foods for prevention and treatment of malnutrition in low-resource settings”

successfully submitted and discussed in 2014 at the University College London (Institute for Global Health), University of London.

Supervision: Andrew Seal and Andrew Tomkins.

Financial support: Valid Nutrition

Ethical approval granted by KEMRI (SSC No. 1414) and UCL


Managing child and adult undernutrition is a global public health priority. In poor settings, improved Ready-to-Use Foods (RUFs) are needed for treatment and prevention of undernutrition.


To develop a method for the design and pre-clinical testing of novel, low-cost, RUF products.

A method was developed and tested. Four sequential studies involved HIV-positive Kenyan adults with severe acute malnutrition.

  1. A qualitative study explored adherence and consumption barriers with the current UN-standard peanut/milk-powder-based formulation (P-RUF).
  1. A study using Linear Programming (LP) designed an improved, cheaper RUF formulation soy/maize/sorghum-based (SMS-RUF).  The formulation was considered accurate if its manufactured prototype, compared to the calculated values, had: a measured energy density difference (EDD) <10%; a protein or lipid difference (P/LD) <5g/100g.
  1. An acceptability study (cross-over design; washout: one-week) compared the use of SMS-RUTF against P-RUF (n=41), using eighteen consumption/safety/preference criteria.
  1. Based on a literature review (28 randomized controlled trials of micronutrient supplementation; outcomes: increased survival and CD4 cell count, reduced viral load), four criteria to determine micronutrient specifications for the SMS-RUF fortification were developed and applied.


  1. The reported compliance with the prescribed RUTF was relatively low, and informed the necessary formulation improvements.
  1. The LP-determined formulation was accurate (EDD: 7%; PD and LD: 2.3 and 1.0g/100g).
  1. The LP-based prototype was acceptable and safe, in spite of the higher average frequency of nausea and vomiting events.
  1. The existing evidence for determining micronutrient specifications for SMS-RUTF posed some challenges for the development of manufacturing specifications. Twelve of the specifications developed for micronutrients for the SMS-RUTF fortificant premix were equivalent to the UN minimum standards; eleven were 2 to 10 times higher.


The proposed set of methods can be used to design and pre-clinically test improved and cheaper RUF products. However, novel formulations should be clinically trialled before widespread use.

Homa Bay hospital (photo by Bannon)

Acceptability and safety trial

Follow-up of this research:

The following papers were written based on my thesis:

The MSF-Scientific day (2008) hosted one presentation . The abstract  is also available on line.

A clinical trial to test the SMS-RUTFh is due to start in the following months.

More publications on related topics are available on a page of this blog.

Interested in the topic of of my PhD? Do not hesitate to contact me for any query.  On the university website, the entire thesis will be available for downloading soon after its discussion with the PhD external examiners.

Interviewing and thanking few of the study participants at the end of the trial.

~ ~ ~

In this page, the pictures were authorized by the study participants.  For more pictures by my  friend Brendan Bannon click here.

– – –

NB – To follow up any further development of my thesis (or any post of this blog), enter your email in the rectangle at the bottom/right side of this page (you can un-subscribe any time)

Open Source: a spread sheet application for planning, calculating and monitoring the Nutritional Value of food

In Under-nutrition on August 11, 2012 at 11:49 am

The planning, calculation, and monitoring application for food assistance programmes, NutVal 3.0 has an expanded database of commodities and products, and new population sub-groups to use for asssessing the adequacy of food assistance. NutVal is designed to run on Excel 2003 and later versions.

Download the most recent version of NutVal

NutVal was developed UNHCR, WFP, IGH/UCL and Global Nutrition Cluster.

– – –

This blog hosts other posts related to the use of nutritional software.

%d bloggers like this: