QUALITY CONTROL AND COST MANAGEMENT SYSTEM FOR CEMENTATIONS

MIXTURES

Background of the Invention

This invention relates to quality control and cost

management system for cementations mixture that uses a closed loop process for reducing concrete performance variability as driven by the production batching process' variance, and thereby significantly improving both quality and economics of concrete.

Prior Art

Concrete performance is generally specified and used on the basis of its 28 day compressive strength, or at times for pavement construction on the basis of its flexure strength at a specified age such as 7 or 28 days. The methods of measurement and reporting are generally specified by ASTM (such as ASTM C39 and C78) and the equivalent International standards such as applicable EN (European Norms) . Additionally, concrete mix design and quality evaluation is guided by ACI 318 (American Concrete Institute) as a recommended procedure, which is almost always mandated by project specifications in the US, and also used in many countries worldwide. In ACI 318 a set of statistical criteria are established that relate concrete mix design strength, F'cr, to it structural grade strength, F'c, as used in the design process by the structural engineer. Thus the concrete producer designs his mixes to meet certain F'cr values in order to meet certain desired F'c structural grades specified in the project specifications. A variable relating F'cr and F'c is the standard deviation of strength testing, SDT, as

determined per prescribed ACI procedures . These ACI formulae are :

For F'c < 5,000 psi :

F'cr = F'c + 1.34 SDT (1% probability that the run average of 3 consecutive tests are below F'c) [1]

F'cr = F'c - 500 + 2.33 SDT (1% probability that a single test is 500 psi or more below F'c) [2]

For F'c > 5,000 psi - [1] applies but [2] is replaced by [3] :

F'cr = F'c - O.lF'c + 2.33 SDT (1% probability that a single test is 10% of F'c or more below F'c) [3]

In general the above equations can be expressed in the following form: Mix Design Strength (F'cr) = Structural Grade Strength (F'c) + An overdesign factor proportional to the Standard Deviation of testing, SDT .

The factor SDT is a direct measure of concrete quality and reliability, and experience shows that it can range widely from an excellent level of on the order of 80 to 200 psi, to the very poor level of over 1,000 psi. Concrete mix design cost factor is directly proportional to SDT, which means that high quality concrete is also less expensive to produce since it would contain less cement (or cementitious materials, which include binders such as slag, fly ash, or silica fume in addition to cement) .

Because of the above ACI approach now in practice for many decades, the industry (ready mix producers, test labs,

contractor, and specifying engineers) has paid a lot of

attention to test results variability and the standard deviation of testing, and has generally speculated and litigated as to the sources of such variability.

Summary of the Invention

A novel feature of this invention is to establish a defined process for quantifying, managing, and optimizing the

contribution of concrete manufacturing (the production batching process) variability, as measured and quantified by the standard deviation SD(AS), in the real time to the testing variability SDT, and to establish real time, software-based cost indices that quantify the SD (AS) contribution to SDT.

One of the novel aspects of this invention is a real time process for linking batching variability (as compared to mix design baselines) to concrete strength performance variability, and interpreting such batching variabilities in terms

performance quality and cost changes. As further discussed below, the real time results are displayed in a summary format so that concrete businesses can at a glance see the quality and cost factors associated with inefficiencies in their operational production processes. These factors are displayed across all of the business' plants, and also as averages for all the plants.

The invention also includes a process for improving

operational efficiency and thereby increasing profitability by 10% to 40%. The algorithm, which can bee implemented through software on a computer, does this by 1) Designating the plant with the least batching variability as the "benchmark plant"; 2) Computing savings if all the remaining plants were improved to the level of the benchmark plant; 3) Through drill-downs to details included in a set of dashboard charts, allowing the user to determine the root causes of the variabilities so that improvements can be made. The benchmark plant approach is important, since it allows the business to set realistically achievable and not blue-sky goals.

The present Invention is a quality control management system for use in the ready-mix concrete industry that is specifically adapted for tens to tens of thousands of concrete mixes that are produced from a few to hundreds of production batching plants, which are providing batched concrete from these mixes to many projects, all managed from a centralized,

consolidated database that can connect remotely to batch plant batching computers or dispatch systems.

The quality control and cost management system has the following attributes:

A database containing all the mix design recipes, test data, and the SDT factors

A way of computing for concrete the cost of the

cementitious materials ($CM) in mix designs for each given structural grade of concrete as a function of the applicable SDT factor by Eq [9] , as detailed in this application.

A closed loop process, as disclosed for reconciling batched concrete component materials quantities and their exact types to the parent mix design recipe components obtained from a master mix design list in the database. The quality control and cost management system can be defined as comprising: a database module having stored therein a concrete recipe, a first tolerance and a second tolerance, the first tolerance associated with an informational alert and the second tolerance associates with an actionable alert; an input module in communication with the database module and transmitting the concrete recipes, the first tolerance and the second tolerance to the database module; a production module in communication with the database module, the production module associated with a concrete

production facility that makes a concrete mixture based on the concrete recipe and transmitting the concrete mixture to the database module; a comparative module in communication with the database modules, comparing the concrete mixture to the concrete recipe, determining if the concrete mixture meets or exceeds the first tolerance and determining if the concrete mixture meets or exceeds the second tolerance; and an alerts module in communication with the comparative module, generating the informational alerts if the concrete mixture meets or exceeds the first tolerance, and generating the actionable alert if the concrete mixture meets or exceeds the second tolerance.

Suitably, the database module, the comparative module and the alerts module are all housed in a single, master module at a single location. Suitably, the single master module is a first computer processing unit at a first location.

Suitably, the input module is housed in a second computer at a second location.

Suitably, the production module is housed in a third computer at the concrete ready-mix facility.

Suitably, there is a single database module which houses the plurality of concrete recipes, a plurality of first

tolerances and a plurality of second tolerances.

Suitably, there are a plurality of production modules each of which is associated with a different production facility.

Suitably, there is a single comparative module and a single alerts module.

Suitably, each one of the modules is a computer.

The quality control management system employs a quality control management method comprising: inputting to a database a concrete recipe, a first tolerance for generating an informational alert and a second tolerance for generating an actionable alert; making a concrete mixture based on the concrete recipe; inputting to the database the concrete mixture; comparing the concrete mixture to the concrete recipe; determining if the concrete mixture meets or exceeds the first tolerance; determining if the concrete mixture meets or exceeds the second tolerance; generating the informational alert if the concrete mixture meets or exceeds the first tolerance; and generating the actionable alert if the concrete mixture meets of exceeds the second tolerance.

Suitably, the concrete recipe includes detailed specifics about proposed ingredients, proposed amounts of the proposed ingredients, and proposed cost of the proposed ingredients.

Suitably, the concrete mixture includes detailed specifics about actual ingredients, actual amounts of the actual

ingredients and actual costs of the actual ingredients. Suitably, the comparing step comprises comparing the actual ingredients to the proposed ingredients, comparing the actual amounts of the actual ingredients to the proposed amounts of the proposed ingredients, and comparing the actual costs of the actual ingredients to the proposed cost of the proposed

ingredients .

Brief Description of the Drawings

These and other aspects of the present Invention will be more fully understood by reference to one of the following drawings .

Figure 1 is an overview of the quality control

management system;

Figure 2 illustrates a closed loop process for reconciling batched concrete material quantities and costs to parent mix designs;

Figure 3 illustrates a closed loop process for proactive management of concrete quality and cost, leading digits indicate different physical locations;

Figure 4 illustrates a closed loop process with truck

added water; Figure 5 is a histogram of real time data for batched

water; and

Figure 6 is a control chart of real time data for batched cementations material .

Detailed Description of the Invention

Figure 1 illustrates quality control management system 10 which has database module 12 that houses cement recipe 14.

Cement recipe 14 are inputted to database module 12 by input module 16. Production module 18 is located at a ready-mix concrete facility which has been provided with concrete recipe 14. During the manufacture of batches of concrete recipe 14 different concrete mixtures 20 will be generated from batch to batch. Each concrete mixture 20 is inputted by production module 18 to database module 12. The concrete recipe 14 and concrete mixture 20 are then outputted along with first

tolerance 15a and second tolerance 15b to comparative module 22 for comparison. First input tolerance 15a and second tolerance 15b are inputted to data module 12 through input module 16. Comparative module 20 compares the concrete recipes 14 with concrete mixtures 20. If concrete mixture 20 meets or exceeds first tolerance 15a then alert module 24 sends out an

information alert 26. If concrete mixture 20 meets or exceeds second tolerance 15b then alert module 24 sends out actionable alert 28.

Suitably, master module 30 houses database module 12, comparative module 22 and alert module 24.

Turning to Figure 2, Figure 2 illustrates the closed loop process. The master list of mixes exists in the database 41, and has been obtained using mix data entry input module 31. A production module 51 which outputs the component types and quantities of batched concrete mixtures 52. A software service that takes the batch data 52 and automatically imports them into the database 41 using a software service program (a program that always runs) . A comparator module 42 that looks up the parent mix design of each and every batched concrete mixture 52, and matches 42b each and every batched material component to its mix design recipe components. A calculation module 43b that per details 43b, for each and every batched material component X, computes and saves to database or to computer memory, wherein, ΔΧ = (Xb - Xm) /Xm, with Xb designating the batch & Xm the mix amount .

For instance, if we set X = C (cement), and the amount of C in the mix design is 500 lbs/cyd, but if the batched amount is Cb = 550 Lbs/cyd, then AC = 10%. If the parent mix design cannot be found, or any batch components cannot by matched to mix design recipe, then a mismatch alert is generated and the calculation is skipped.

A closed loop process, as illustrated in Fig. 2 of

drawings, for reconciling batched concrete cost of materials to the parent mix design cost components obtained from a master mix design list in the database, illustrates the master list of mixes exists in the database 41, and has been obtained using mix data entry input module 31; material component types and unit costs 22 that are inputted into the database 41 using input module 31; a production module 51 which outputs the component types and quantities of batched concrete mixtures 52; a software service that takes the batched data 52 and automatically imports them into database 41 using a software service program (a program that always runs) ; a comparator module 42 that looks up the parent mix design of each and every batched concrete mixture 52, and matches 42 each and every batched material component to its mix design recipe component; a calculation module 43, that per details 43c, for each and every batched material component X, computes and saves to database or to computer memory:

Δ$Χ = ($Xb - $Xm)/$Xm, with $Xb designating the batch & $Xm the mix component cost, and where $X for component X is its amount X in the batch or mix design multiplied by component X's unit cost; for instance if we set X = C (Cement), and the amount of C in the mix design is 500 lbs/cyd, and its unit cost $100/ton, then $Cm = 500 x 100/2000 = $25/cyd, but if the batched amount is Cb = 550 Lbs/cyd, then $Cb = $27.5/cyd if the parent mix design cannot be found, or any batch components cannot by matched to mix design recipe, then a mismatch alert is generated and the calculation is skipped.

A way of uploading mix recipes to batch plant

computers or dispatch systems as per process as illustrated in Fig. 3. This is in order to establish proactively managed mix design target values for the production batching process as the forward loop 47 of the closed loop process, where the first digits of the identifying process components indicates a specific physical location; the steps 21 & 22 occur at one location, and the steps with the leading digit 4 at a different location, and the ones with leading digit 5 at yet another location; the forward loop 47 is responsible for single or mass uploading of mixes and materials to the production system remotely, and securely via the internet; a mix export module 44 optimizes mixes for cost and technical performance, and selects a subset of mixes and materials from the database for uploading; a FTP file transfer service now takes the upload data and via the Internet or the Intranet makes the data available at

location 5 for importing into batch plant computers (also referred to a batch panels) , or into a dispatch systems

database; the forward loop of the closed process thus makes possible establishing target mix designs or formulations

proactively at multiple remote locations, via the Internet or Intranet .

A way of retrieving batch results, as per process detailed in Fig 3, particularly as concerning the plant level water (WP) and cementitious (CM) amounts in the real time and automatically on an on-going basis in order to reconcile the batch results to their target mix design factors as the return loop (55) of the closed loop process. Figure 3 illustrates closed loop process for proactive management of concrete quality and cost; leading digits indicate different physical locations. Forward loop uploads mixes to establish batch targets, and return loop retrieves batches to reconcile to mix target.

Data sets 52 for batched concrete mixtures are

automatically exported from batch panels or dispatch systems using established industry practices. A software service program 53 now transmits the retrieved data 54 via the Internet or Intranet to location 4 where the mix design master list resides in database 41. The data 54 are imported automatically into the database 41 by software service 411.

As illustrated in Fig. 4, a way of retrieving batch water additions (WTR) during transportation in concrete trucks

equipped with automatic water metering devices and transmission of water additions to a cloud database in the real time, and obtaining such data from the cloud database automatically on an on-going basis in order to reconcile the total water batched results (WP+ WTR) to the mix design water target factor as the return loop of the closed loop process for water reconciliation.

In Figure 4 is shown the return loop of added truck (WTR) water; the purpose is to add WTR to water added at the plant and then reconcile to the mix design water. A concrete truck at location 6 transmits via the Internet or Intranet to a cloud database 70 the added water to batched concrete during

transportation .

A software service 71 exports the added water amounts for many trucks for a given project via the Internet or Intranet to mixture design database 41. A software service 411 receives this added water data, and imports it automatically into the mixture design database 41. As discussed above and per details of the closed loop process in Figs 2&3, a way of computing the quantity differences between batch results and mix targets and expressing them as percentages or lbs (KG) , which are computed per below formulae for the all important cementitious and water amounts for any given mix:

ACM = (Batched CM for a given Mix) - (CM amount in the Mix design)

AW = (Batched W for a given Mix) - (W amount in the Mix design)

Where W = total batched water for a single truck load and could include both the plant added amount of WP and the truck added amount of WTR. However, since in general the industry as a matter of practice holds back a small fraction of the target mix amount as the so-called "trim water" of around 5% to 10% of the mix water, in many cases for variance analysis the WTR amount can be neglected with minor effects on the batch water variance.

Having now accumulated all the needed batched concrete and mix design data per processes detailed in Figs. 2,3,& 4, and having computed the all important ACM/CM and AW/W , a way of computing the respective standard deviations of SDrCM (r means relative, as in the batch relative to mix design amount) and SDrW from ACM/CM and AW/W, for each concrete batching plant and across all its production batches and mixes; where: SDrCM = Standard Deviation of ACM/CM = Standard Deviation of the differences of batched versus mix design CM for all batches produced in given plant for a selected date range,

And, where:

SDrW = Standard Deviation of AW/W = Standard Deviation of the differences of batched versus mix design W for all batches produced in given plant for a selected date range such standard deviations as needed above are computed by standard statistical methods.

Then noting that per principles of concrete technology, and since strength is proportional to CM/W ratio, it can be shown that for any given mix:

(AS/S) due to batching variability = (ACM/CM) - (AW/W)

Which means that for each given mix design, there is a relative strength gain with increasing CM amount from the mix design baseline; and a relative strength loss with increasing W amount from the mix design baseline.

Per principles of statistics (VAR = Variance) :

VAR(AS/S) = VAR (ACM/CM) + VAR(AW/W) . [4]

2

= (SDrCM) + ( SDrW ) [5] Let :

SDrWCM = the standard deviation of the ratio W/CM relative to the mix design W/CM,

Then:

(SDrWCM) = [(SDrCM) ² + (SDrW) ² ] ^1/2 [6]

Hence: SDrS = (SDrWCM) [7]

Where :

SDrS = Standard deviation of relative strength as brought about by the variability of the batching process. By "relative strength" is meant strength changes from the mix design

baselines for all batches in a given plant due to the batching variabilities of CM and W from their respective mix design baselines, expressed as a ratio to their mix design baseline strengths .

It follows that:

SD(AS) = S x (SDrWCM) [8]

The closed loop process uniquely reports in the real time the batching statistical factors SDrCM and SDrW, from which, in the real time, the statistic SD(AS) is computed per Eqs [6],

[7], and [8] . SD(AS) is a direct measure of concrete strength performance quality as affected by the quality of the production batching process, both of which are characterized by the

applicable SD values. Low quality is defined by a high SD value, and vice versa, high quality by a low SD . Thus Eq. [8] basically states that as the batching quality deteriorates, so does the strength quality in a proportionate way.

It follows that when the batching quality decreases, it's necessary to adjust the applicable mix designs by using an extra batching driven increment in the SDT standard deviation factor. This is done using the ACI318 Eq. [1-3] plus Eq. [8] in the

following form:

AF'cr = 1.34 x S x (SDrWCM) [la]

AF'cr = 2.33 x S x (SDrWCM) [2a]

AF'cr = 2.33 x S x (SDrWCM) [3a]

Where in each case, AF'cr is the added mix design strength increment resulting from the batching variability SDrWCM, for each of the three ACI equations. Since [2a] and [3a] are

identical, the three ACI statistical criteria are in fact reduced to two for these batching increment cases.

Remembering that F'cr is the mix design strength,

increasing it means increasing the CM content at constant W, and thus increasing the CM cost in the mix. This CM cost of a mix can be quantified as follows:

Φ = CM efficiency factor in PSI/ (LB.CYD)

K = CM cost per LB

$CM = CM cost per cyd = (Κ/Φ) x F'cr [9]

Now using Eq. [la-lb] plus [9] :

A$CMB = increase in CM cost due to batching SD

A$CMB = 1.34 x (Κ/Φ) x S x SDrWCM [lb]

ACSTB = 2.33 x (Κ/Φ) x S x SDrWCM [2b] or [3b]

These and other aspects of the present Invention may be more fully understood by the following Examples.

Example 1

Illustration of the Impact of Concrete SD on its CM Cost

1. The table below is generated to illustrate that concrete variability impacts its CM (cementitious cost) cost very significantly .

2. The analysis is performed for a concrete of structural grade 4,000 psi, and using the referenced equations previously derived in this document. The example analysis assumes a CM efficiency factor, Φ = 8 psi/ (LB . cyd) , and a CM cost, K = $0.045/Lb.

Starting at a SD of 200 psi, the SD is increased in 100 psi increments in column 2, the mix design strength computed in columns 3 & 4 per two different ACI formulae, with the higher value always governing.

The mix CM cost is computed in column 5.

The cost of quality variability is well illustrated in columns 6 & 7 ; in column 6 it's shown that per each 100 psi increase in standard deviation of strength, the CM cost will increase between $0.75 to $1.31 per cyd. In column 7 is shown that the CM cost relative to very high quality concrete (row 1) can increase dramatically by more than $8/cyd. Noting that the concrete industry on average generates a net profit of on the order of $0.5 to $2 per cyd, this example (uses realistic numbers) illustrates the tremendous importance of maintaining low variability.

An important factor for maintaining low strength performance variability is the consistency of the batching process, the theory for whose influence has been formulated in this

document . Ref# 1 2 3 4 5 6 7

Eng Design Mix Design $CM Relative cost of

Strength Strength: F'cr, psi $CM/CYD per 100 psi SD Variance

Ref# F'c, psi SD, psi Eq [1] Eq [2] Eq [9] DEL_$CM/cyd

1 4,000 200 4,268 3,966 $24.01 $0.00 $0.00

2 4,000 300 4,402 4,199 $24.76 $0.75 $0.75

3 4,000 400 4,536 4,432 $25.52 $0.75 $1.51

4 4,000 500 4,670 4,665 $26.27 $0.75 $2.26

5 4,000 600 4,804 4,898 $27.55 $1.28 $3.54

6 4,000 700 4,938 5,131 $28.86 $1.31 $4.85

7 4,000 800 5,072 5,364 $30.17 $1.31 $6.17

8 4,000 900 5,206 5,597 $31.48 $1.31 $7.48

9 4,000 1,000 5,340 5,830 $32.79 $1.31 $8.79

Next some real time batch data variability with respect to the mix design factors is reviewed, and then such data are used for an example quantification of the cost of mix strength

performance variably as driven by batching variability.

Example 2

Typical Batch Variability Data & Calculation of the Standard Deviation of W/CM relative to its Mix Design value

Below are included three example data sets obtained from the closed loop process in the real time.

1. In Fig. 5 is shown the Histogram of real time data for

Delta_Relative Batched Water (W) : +1% means that the amount of batched water is 1% higher than that prescribed by the mix design. Limits show the ASTM C94 Limits, and labeled "A" and "B" . 2. In Fig. 6 is shown the Control Chart of real time data for Delta_Relative Batched Cementitious (CM) : +1% means that the amount of batched CM is 1% higher than that prescribed by the mix design. C and D show the +/- three SD Limits.

3. In Table 3 below is shown a closed loop real time report for batching standard deviation for CM and W relative to the mix design baselines, and shown computed are the standard

deviations of both W/CM and Strength; for this operation, plant 402 would be the benchmark plant at a strength SD of 140 psi; plant 223 at 609 psi would be more than 450 psi above this plant which implies an additional cost of about $6/cyd compared to the benchmark! Compared to a net profit of $0.5 to $2 per cyd, this is a huge cost.

Table [1] Example Quantification of Batching Data Variability

1. Table [1] below includes a set of actual real time data in columns 1-5

2. Column 6 shows the computed standard deviation W/CM as

computed by Eq [6] and using the raw data from columns 3 & 5.

3. Plant #141, row 9, is designated as the benchmark plant since it shows the least variability

Table [2] Example Quantification of Strength Standard Deviation due to Batching Variability, and the Resulting Cost

1. Assuming an average concrete mix design strength of 4,000 psi, shown in Table [2] are the strength SD (Column 3) computed from the SD of W/Cm; the strength SD varies by more than a factor of 5 from 85 psi for the benchmark plant, to 458 psi in plant #124.

2. If this batching strength SD could be reduced to the benchmark value, then significant CM costs could be saved as shown in column 4; this cost factor varies from $0.02 per cyd to $2.85 due to the varying batching qualities of the plants.

Another way of looking at this situation is to suppose that the mix designs developed for the benchmark plant are used across all the plants. This could lead to a very costly situation since probability analysis shows that for each 100 psi increase in strength SD from its assumed mix design value, the failure rate will increase by more that 4%, which

translates to a potential remedial cost of around $2/cyd per 100 psi of SD increase.

Table 3 - Closed Loop W/CM Ratio & Batching Strength Standard Deviations (last two columns) from real time data in the other columns.