Concrete Mix Alignment

It's critical for the concrete supplier to manage their mix designs if they want to be successful. Mixes must not only meet the specified performance requirements, but they must be economical, and they must be something a contractor looks forward to using. A producer’s mix portfolio can quickly grow and evolve to become seemingly unmanageable. Not only can the number of mixes increase exponentially as one examines all of the design factors (such as strength, stone size, slump, exposure class, and so forth), but over time with day-to-day adjustments these mixes can quickly grow out of “alignment”. As such, it is critical that the concrete producer has a strategy in place to manage their mix designs and there are a number of factors to consider.

Standard Mix Coding System

A well planned, comprehensive mix coding system is critical to managing one’s mix portfolio. The mix code should quickly and easily identify what the particular mix is designed to do and the important factors going into the mix design. Although it can be challenging to come up with a system when limited characters are available, this will improve communication between QC, batchers, drivers, customer service, and even your customers.

I’ve gone into a number of facilities where the coding system only serves to create confusion. They might have “mix 781” and “mix 334”, or they might have something like “32 MPa sulphate” (to what degree?) or “John’s Driveway Mix”. Not only do these not say a whole lot about your mix, but with a poorly planned system you greatly increase the risk that the wrong mix will be sent. A good coding system will break down the various input and performance factors that go into a concrete mix design and allow you to know exactly what you’re doing with a quick glance.

For example, your first letter might denote the application type such as “R” for residential, “M” for municipal, “E” for high-early, “P” for pumping, “S” for standard, and so forth. This can then be followed up with two numbers for the strength, and then another two digits for the exposure class. You can then pull the appropriate digit from each column and then simply describe your mix. For example, a “M32C2” would be a 32 MPa, C2 exposure class for municipal work. The rest of the digits can then be used for extras and value-added products, such as SCM replacement levels, maximum aggregate size, slump targets, admixtures and additives, and so forth for what makes sense for your business. So you could have something like “M32C2CMLS” which could mean a 32-C2 city mix for Curb Machine with Crushed Limestone.

Of course, this is somewhat of a simplification and it will vary based on your own market and governing specifications, but hopefully you get the idea. You want to capture the critical elements and follow some sort of structured hierarchy from the most important information to the less important information, to the optional and extra factors. You want to keep it easy to understand to avoid ambiguity and confusion, and need to have the forethought when designing it that it will be flexible enough to accommodate new mixes or considerations.

On top of this, it is important to manage the number of mixes you have. A mix portfolio list can quickly grow to seven or eight hundred mixes if you don’t stay on top of things. Realistically, depending on your market and capabilities, you can probably prune this down to maybe two or three hundred if you get rid of duplicates, or obsolete old custom mixes. What’s really neat is that the 80-20 rule applies quite literally to this – 80% of your volume is likely in your top 20 mixes. And of course, these are the ones we need to focus on most overall with our whole alignment approach.

Concrete Mix Design Alignment

“Mix Design Alignment” means that we design our concrete mixes with an approach of viewing concrete as a spectrum instead of evaluating things on a purely case-by-case basis (although exceptions will exist in individual cases).

What do I mean by this? I mean that when we look at all of our mixes together we should have some very clear relationships between the performance requirements and the material proportions. And these relationships should follow a simple straight or more likely curve-linear relationship. In my experience, these relationships (if you have them right) hold from zero-slump concrete, to conventional mixes, all the way to SCC.

If you follow a standard mix design process, like those laid out in ACI 211 for example, you will already have these relationships built in to a good degree. You have relationships between aggregate size and the stone, water, and air contents, between strength and durability and w/cm ratio, and so forth. However, I like to build in other relationships, particularly with the stone content and the slump (or slump flow) as it has a significant impact on the workability and cohesivity of a mix, and with the water (or water reducer) content and w/cm as the paste thickness or viscosity also has an impact on workability. If you plot a performance requirement like slump or strength, you should see some clear trends with some of your mix inputs. Although there will be some exceptions, if you see a higher strength mix with a higher w/cm ratio than a lower strength mix for example… well... you’re out of alignment.

Following a standard process such as this means that we’ve developed tables of relationships between our mix design performance parameters and mix design proportions (and hopefully have these tables written down somewhere!). Being standardised and simplified, it is easier to make broad adjustments to the entire mix portfolio based on adjusting these relationships as our raw materials or other circumstances changes. If everything is on a straight line we can simply adjust the relationships and “move the whole line up or down”. On top of this, it provides a nice tool for quickly interpolating or extrapolating from that data to design new mixes if the need arises.

Performance Review and Change Log

Concrete performance should be periodically reviewed to find areas of performance concern and opportunities to economise our mixes (winter is a great time for that up here in Canada!). Generally you want to focus on your big hitters such as your top ten or top twenty mixes by volume (but at the same time you don’t want to ignore your other mixes). If our mixes are aligned, it’s much easier to make appropriate adjustments. Typically we want to balance our performance and our cost, and will draw on statistical performance, what we see in the field, and our theory and knowledge (and hoping that all three point us in the same direction).

One particular chart that I absolutely love (because it’s so simple, but tells you so much) is to graph our actual values against our design values for the parameters of interest, over our whole range of mixes. If you set this up as a proper Table in Excel, this provides the added benefit of being able to filter through the data. For example, if we plot actual strengths against our design strengths from 15 MPa up to 50 MPa we should see a very clear trend with actual strengths increasing. If we see something funny, like a 30 MPa mix having higher strength than a 35 MPa mix, we can drill down and see what’s going on by filtering by mix code, slump, air content, plant, date, or any other factor which we are recording. Just as our design inputs correlate with our performance requirements, our performance results should correlate with the performance requirements as well. This may sound obtuse, but it’s often the simple things we fail to validate.

(How can you not love this chart? A quick glance tells you exactly where you stand!)

Likewise, simple run charts (or more properly Control Charts) provide valuable insight into time related trends and possible sources of variation or cause for concern. This can be a bit difficult when we have numerous mix designs and parameters, but sometimes we can combine a number of different mixes by tracking the difference between the actual result and the target on one graph. Additionally, simple x-y plots can tell us a lot of information… is your strength decreasing as the slump target increases? Then perhaps you need a little less water and a little more admixture as the slump increases.

When making mix changes, I like to make realistic adjustments, keeping in mind our preferred risk level. It’s better to make a conservative adjustment when trying to economise mixes and to re-evaluate again a month or two later. One critical component of a Quality Management System is a Change Log. You need to document the date, what was changed (original and new value), a reason, and what the result was. Unless you have a perfect memory (and no… you probably don’t) ignoring this part would be like burning a road map behind you as you travel - a surefire way to get yourself lost. This is one of the most overlooked but simplest things we can do to give ourselves insight into our product, process, and performance.

Case Study

A number of years ago, I was asked by one customer who was looking to reinvent themselves, to come in and help redesign their QC system from the ground up. Their previous focus was “as cheap as possible” with no real management system and a shoot-from-the-hip approach with reacting to problems instead of proactively managing their business. Some mixes were failing regularly (some over 40% of the time!), while others were nearly twice as strong as they needed to be. There were a number of duplicate mixes where, for example, one wall mix might have 1000kg of stone, while another had 1150kg of stone and other things that didn’t make intuitive sense when you looked at the big picture. Air factors and yields were all over the place. Rip out and repair costs were almost sinking them, and of course… there was no documentation to be found anywhere on what was going on. Basically, the root cause for their failing business was simply a lack of system.

We started with an overall vision of quality and process management forming the foundation upon which customer focus and business growth would develop. Naturally, this involves standardisation, planning and proper design, identifying and managing sources of variation, monitoring and control, documentation and review, and training and communication. We built a system in place built around these fundamental elements.

We had to address the mix designs as well and basically… we wiped out the 800+ mixes they had and started fresh with the main twenty they would need which were built healthy. It took a good few months to get everything in place, and of course embraced continuous improvement and evolutionary operation. By following a structured process to implement a management system, we went from a 28% failure rate to less than 2%.

There were some very strange side-effects that came along with this improvement – customer feedback was improved and they renewed their image and reputation, and although there were very slight cost increases on some of the mix designs, the mixes were “in demand”. Additionally operational efficiencies increased, particularly with time spent “fixing a load”, which resulted in reduced costs overall. Who’da thunk it? A quality product isn’t a cost center… it’s a vehicle for a sustainable business strategy!

Summary

Having a standardised and aligned mix portfolio helps make our products more manageable. There are relationships between concrete performance, cost, and mix design proportions and these things should make sense! Our products should be properly designed, based on the requirements and a safety factor to accommodate variation. At the same time we want to anticipate and eliminate variation from our process and ensure that we have standard operating procedures and clear lines of communication. Monitoring and routine evaluation of product performance is key to identifying sources of variation and for mix optimisation to increase the competitiveness of your company. Remember, Quality Control is not just about running slumps, but about Process Engagement and everyone has their part to play!