WHAT ARE YOU LOOKING FOR?
Credit : GMP +
As for human gastronomy, the performance of the feed highly depends on the quality of the ingredients we are selecting. Such selection depends on our ability to analyse each ingredient contents. It requires repetition of analysis batches after batches. But it is expensive and we quickly arrive to limitations. To be efficient, it is critical to set a plan to know in advance what we need to analyse. It would be too expensive to analyse everything. The smarter feed millers are the ones who know what to analyse to allocate most of their budget on it according to a deep analysis of their nutrient risks.
In previous decades the implementation of QM systems in the feed milling industry have revolutionized production technologies and dependent processes, leading to high product quality standards and product reliability state-of-the-art. Last but not least QM systems make a large contribution to a positive reputation of the feed industry within the food chain. High quality in feed production is based on reliable determination of the composition of raw materials and end products. Feed is produced batch-wise and except for water and oil, there are no possibilities to adjust the composition during production. Hence feed quality and control mainly depends on the measurement and managing of all raw materials before they should be released for production. Despite often observed high quantity of end product analysis, such detected nutrient deviances do not have a useful impact on traceability and correction of failed product quality. Nutrient respectively constituent analysis management of feedstuff has become a fixed part in the business process and is considered to be mandatory. It is in hands of the quality manager to establish a sampling- and testing program in which he has to line out the frequency of sampling and the analysis pattern for each feedstuff.
However, any extra cost arising therefrom cannot be covered by simple adjustment of feed sales prices. Thus, besides being precise, product quality control has to be fast and notably cheap. Not unusual, the pressure of cost lead to management consideration of minimizing in-house laboratory activities and it can often be noticed that activities are shut down by outsourcing such services. Despite such reorganization measures, they often fail to achieve the desired cost savings and end up reducing scope and sequence of analysis plans maintaining a weak QM system afterwards. Preventing such negative developments, main focus should be laid on the quality of the sampling plan for nutrient analysis of feedstuff, as analysis represent the highest volume and imply the major cost factor in lab quality control. Being effective, optimization of the sampling plan should accomplish its major obligations:
Creating an effective system in order to identify quickly enough major variations between raw materials batch received and the formulation matrix
Reducing the number of analysis, aim to reduce the total analysis cost significantly.
Technically, an effective sampling plan should be based on:
Major risks of raw materials variations - nutrient variability is higher for Meat & Bone meal, Coprah meal and Rice bran than the other raw materials
Contribution of each raw materials to each formula nutrient – it is not about how much quantity of one ingredient we incorporate into a formula but rather how much one ingredient influence one nutrient – for example – analysis show that fish meal is responsible for 89% of methionine variation in formulas
How critical is a nutrient - energy, digestible amino-acids
Requirements of QM systems, e.g. ISO, GMP, HACCP
In a lot of cases, analysis for “Starch” is missing although it is an important parameter in energy calculation formula (cf Newsletter May 2015 – NET IS BETTER THAN GROSS) but also characterized by high variation factor, e.g. in cereals.
The graphs below are the results of 56,000 samples made in South East Asia by Evonick measuring the variability of amino acids content of several raw materials. The results are expressed in % of coefficient variation. We consider that above a CV% of 10, the raw materials is highly variable and a special attention should be given in the analysis of these raw materials at reception especially for raw materials whose contribution to the formulas is high.
For ingredients as poultry meal, fishmeal, meat meals and rice bran, nearly all amino acids are highly variable and would require special attention from the QM team. For Wheat bran, Lysine and Arginine are the 2 amino acids to be looked at. In certain raw materials some constituents give no significant contribution to the total energy, consequently there is no need or less need for frequent analysis, e.g. crude fibre. Moreover, because the level of those nutrients is either low, and/or rather constant, it is very ineffective and costly to concentrate analysis activity on such constituents. Such values should be completed using approximate existing in-house analysis results or should be supplemented using relevant figures from approved data bases, e.g. “CVB Feed Table” or “INRA Table of composition and nutritional value of feed materials”. The following table shows a practical analysis pattern of important feedstuff taking these considerations into account.
The table shows that out of total 126 necessary constituents only 74 relevant nutrients need to be analysed. 52 (this counts for 40 % of total constituents) can be reasonably completed by approved approximate figures. Creating an effective system in the selection of specific nutrient analysis for feedstuff can increase the effectiveness of the QM system as such by reducing total analysis cost significantly at the same time! Once a strong QM system is in place, it is important to use the information they produce to contribute to an improvement of the feed performance. Based on the analysis of each batch, the management have several options to handle the non-compliant batches
Rank the supplier according to the CV% of their delivery in order to minimize such problems in the future. We can assess the purchaser performance based on the CV% of the products they are buying and integrate this indicator into the measure of their performance
Negotiate a discount from the supplier to compensate for the non-compliance
Minimize the contribution of variable ingredients into the formulas
Allocate non-compliant batches to less critical formulas
Update the matrix to anticipate raw materials variability (formulate based on nutrient values from lowest batches)
If you are interested in making a thorough analysis of your nutrient risk exposure, our team of experts is available to support you;
please contact me at email@example.com