I assume you mean this as a question, and from that point reconciling the mass balance in an operation or a unit is a way to determine operational performance and to compare it to planned or predicted performance. That being an obvious statement, the important part is comparing the the proper results. At any particular point in time you can look at a measure of the feed into a plant or unit and the product coming out, but they will often have different basis (such as % moisture), and will need to be reconciled to the same basis for accurate comparison. In addition there is a lag of several minutes or more between what is in the feed and what the product is and operating conditions or even feed characteristics may have changed, meaning that what is being produced now is what was fed a while back, and this can even be complicated by several circuits.

To get an accurate picture ti is important to compare everything on the same basis, actual feed time, factoring for changes in moisture and density, and others.

Mass balance concilliation is more important. However, for doing that unit operation evaluation has to be done. So both have their importance in system studies of any plant operation to optimise and improve the operation.
The mass balance is of utmost imporatancia crushing processing flotation milling and these give us a parameter which we can optizar our process and continuous improvement.
If you do not measure it you cannot improve it.

Measuring and improving can be two different things. I know of one operation that spent a fair amount of money every year to have a research organization come in and do a complete set of samples so that the flowsheet balance could be recalculated completely. They actually called it a process audit. The problem was that neither the operation or the research organization ever used the data to point the ways to any improvements. It was generally a waste of money without using some creativity to interpret the work.

Every operation I've come across has some variation of key production statistics where key process numbers like weight recovery, concentrate grade, tailings grade, and key measurements like primary mill feed size are used to determine the overall health of the operation. If a deviation occurs more intensive sampling is done in that area to either define the problem or find a solution. The same applies to key areas like comparing new equiment performance with old equipment to evaluate a total change.

There are also monthly reconciliations where stockpiles are surveyed, shipments are added up, and so on to determine the adjustment needed to make the plant's key numbers match up with what was actually produced.

From an operating standpoint it is very rare when a full metallurgical balance of an operating plant has to be redone. It might be nice to have but like the plant in my first paragraph it is probably not worth the effort in a mature plant.

I agree with all of it. But if I buy 1.5 million tons of iron ore and produce 800,000 tons of steel, pay $175 per ton of iron ore pellets delivered, to be used in a DRI/EAF mill, selling the steel at $550 per ton, where am I losing money, investment and how do I stop it, but fast. That is what this is all about. The analysis, measurements, calcullations are an absolute must, the studies of each piece of equipment and its efficiency and maintenance and possible replacement to a better system is a must and it extends to the civil engineer who wanted to save a buck by not putting in ground barriers or pavement or cement runoff to stop ground contamination that goes down 10 meters after 5 years. And where are the losses, how do we control them and recover the materials and what is the value of the materials at each point of loss or spillage or from poor reduction and smelting. How high is the landfill pile and what fines do you pay for landfilling? Is the pile hazardous, is it leaching into the water table? How big is the bank account to pay for cleanup when the authorities find out.? And finally where is your next job, after you say I don't know to all this?
A small team dedicated to this study, continuously, will allow an ongoing evaluation and development of new processes and as compared to what was sold to you and installing new efficient systems based on new developed technologies. Been there done that.

The mass balance done with original data (throughput, assay, size fraction fraction for cycloning, etc) is generally incoherent and inconsistent due to the errors of the data used (measured or estimated).
The mass balance done with data reconciliation method will find and use the best estimates of the original data as close as possible with them which satisfy the mass conservation law. The minimization of the sum, for all data, of the square of the difference between each original value and its best estimate is the basis of the method.Then, the proper or best estimates of data, reflecting the picture of all available values and impressions, will be generated. They will be used for proper performance results evaluation, as mentioned by Mike Alberecht, and modelling and simulation best estimates parameters will be also determined. By using rigorous reconciliation method every times, the comparison will be done on the same basis and the impact of optimization and improvement will be clearly measured as mentioned.

Yes the full mass balance has to be done at reasonable frequency but the mass balance reconciliation can be done frequently for the entire plant operation or main steps. Could you deeply examine your plant auditing reports to find positive points to use for plant optimization. You can also recommend some improvement on the plant auditing campaign objective, procedure, report results and presentation.
I also agree with you Thomas on the importance of a small team devoted to plant evaluation, study, optimization and development of process.

Victor,

I would like to start off by reiterating what Laurie has said: "If you do not measure it you cannot improve it."

Plant surveys are mostly done on an ad hoc basis and for very good reasons, some of which have already been mentioned.

There are two reasons why one would also like to do daily measurements on a plant:

The first reason is for metal accounting. This is according to my opinion also what most metallurgist and process engineers are familiar and comfortable with. My opinion follows from my own experience when I started working on a flotation plant a few years ago.

The frequency of measurement for metal acounting is of less importance as long as your composite samples are representative of your actual process.

The second reason one would like to measure your process is for process control purposes. This is according to my opinion less understood by process engineers and metallurgists.

Here the frequency of measurement is very important:
* Less so for feed and tailings streams that are less dynamic.

* More so for concentrate streams that are more dynamic.

My suggestion would therefore be to ensure that all metallurgists and process engineers responsible for good recoveries and product grades should have a clear understanding of the difference in order to make sound process decisions.

Paul, I agree your conclusion about the knowledge improvement recommended for the best control of the process. I extend it to the mass balance for metallurgical accounting by recommending the use of the data reconciliation method to improve the mass balance and to determine the best estimates of: the original measured and estimated data (throughputs, compositions, etc); the production results; the recovery.

Further to your discussion about data reconciliation; I can only agree that one should assign relevant error margins to specific flows and assays, then balance the streams through minimizing the sum of the square of the errors. It is just important to understand to which data to assign what error margin (higher typically for larger process streams with lower grades).

It is also a very good start to keep metallurgical lab personnel responsible for collecting and preparing samples motivated! It happens quite often that people responsible for these tasks do not have a full understanding of the implication of short-cuts.

You do very well to remind the importance of giving more motivation or consideration, appreciation and implication to laboratory personnel who is responsible of samples collection and analysis for plant control and metallurgical accounting.
About the mass balance reconciliation, it is really important, as you say, to understand to which data to assign what error margin but the most important thing is to know how to use the maximum available data and their errors margins and submit them to the data reconciliation method and generate the best estimates of: mass balance; data (flows-assay); errors; production results; efficiency - recovery; etc. The detection of anomalies, the problems solving and performance optimization can be done, as says Laurie the optimizaction requires the measurement.

Plant is designed based on lab/pilot plant studies to develop the commercial process flow sheet. In actual practice, the feed characteristic sometimes change which effect the plant performance.Accordingly its necessary to maintain the feed characteristics by suitable planing in the mines. For proper plant control, samples need to be taken at important points and if there are wide variation of the values at different point in comparison to standard, overall system studies of plant need to be done for process control as well as necessary rehabilitation of circuits as may be required.
For me Mass balance reconciliation it is basically the key to have a perfect evaluaton of your plant and you will be able to measure the level of variation of parameters, without that you can't know very well what is your result and the reliability of each equipment( feed rate, interaction, output).

https://www.youtube.com/watch?v=pJDtQGKZuSY

https://www.youtube.com/watch?v=gnFCAMIN4qI