DRAFT

COMPASS USERS GUIDE

This Users Guide has not been released for publication. It is provided on an as-needed
basis for informational and evaluation purposes. No part of this document may be
reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, recording, scanning, or any information storage and retrieval
system. Any reproduction, transmission, or distribution of this document must first be
authorized in writing by both the FWHA and The Transtec Group, Inc.

1. INTRODUCTION
COMPASS stands for Concrete Mixture Performance Analysis System.
COMPASS is a windows-based application system that can be used to optimize the
materials selection and proportions of paving concrete based on job-specific conditions.
To achieve this, COMPASS provides the following capabilities:
Identify relevant performance criteria that are a function of job-specific inputs
such as the importance of the project, type of pavement, climatic conditions,
construction constraints, and environmental exposures.
Identify applicable mixture performance criteria and recommended test methods.
Assess the impact that changes in materials or proportions, environmental
conditions, and construction procedures have on portland cement concrete
(PCC) constructability and performance.
Provide analysis tools for aggregate blending.
Recommend initial mixture proportions.
Optimize mixture proportions based on multiple job-specific criteria such as
strength, cost, and permeability.

1.1 Hardware Requirements

The minimum system required for proper operation of COMPASS is a computer
with 128 megabytes (MB) of memory and a processor of at least 500 MHz. The
recommended system is a computer with 256 MB of memory and a processor of at least
1 GHz. COMPASS is designed for a screen resolution of 1024 by 768.

1.2 Operating System Requirements

COMPASS requires Windows 2000 SP1 or Windows XP Professional.

1.3 Initiating the Program

When COMPASS is first opened, the screen shows a menu for File, Edit, and
Help (Figure 1). These menus are used to work with project files, modify project options
and access help features.

A random tip of the day appears each time COMPASS is started. The tip of the
day offers useful facts about concrete.

1
 DRAFT

Along the left side of the screen there is a menu with the four computer modules
that can be accessed. The computer modules are where the user defines their job-
specific inputs and concrete mixture criteria. Each module has the ability to be used
independently as a stand-alone tool, or the user can perform a comprehensive analysis
using all four modules. Each module builds upon one anothers consecutive modules
analysis. The computer modules are:

1. Mix Expert,
2. Gradation,
3. Proportioning, and
4. Optimization.

Under each module, a list of most recently used (MRU) files is displayed.

Figure 1: COMPASS Starting Screen

If the user decides to run a comprehensive analysis, the following sequence

occurs:
Identify the performance criteria relative (or important) to job and site-
specific conditions. (Module 1 - Mix Expert - Performance Criteria)
Select materials that best suit the conditions. (Module 1 Mix Expert -
Materials Selection)

2
 DRAFT

Determine optimal aggregate gradations for maximum packing density

and optimal workability (Module 2 - Gradation)
Determine preliminary mixture proportions. (Module 3 Proportioning
based on ACI 211)
Optimize the concrete mixture proportions based on job-specific criteria.
(Module 4 - Optimization)

The inputs to the first module are cross-referenced with the information in the
COMPASS Knowledge Base to help guide the user in identifying the mixture
performance criteria and selecting the materials to meet these criteria to optimize the
paving mixture for the environment in which their pavement project will be constructed
and expected to perform. Modules 2 to 4 include analytical subroutines to optimize
materials proportions based on job-specific criteria. The following paragraphs provide a
more detailed description of each module.

2. MIX EXPERT

This module consists of a two-stage analysis where mixture performance criteria

based on site-specific conditions are identified first and then the user selected materials
are evaluated in terms of their effect on the mixture performance criteria of interest.

In the Performance Criteria Analysis, users can select their own mixture design
criteria and/or obtain recommendations on design criteria from COMPASS through the
knowledge base as a function of site-specific conditions.

Recommended optimization criteria include:

Slump,
Set Time,
Air Content,
Strength,
Heat of Hydration
Temperature of Concrete
Coefficient of Thermal Expansion
Shrinkage
Permeability
Freeze-Thaw Resistance
Alkali-Silica Reaction (ASR) Resistance
Sulfate Resistance
Corrosion Resistance
Abrasion Resistance

3
 DRAFT

2.1 General Project Information

To begin the analysis, the general project information can be entered into the
program. This is a screen where the user can identify the project and enter specific
notes about the project.

2.2 Job-Specific Info

The next screen is the Job-Specific Info screen (Figure 2). This screen includes
general job-specific information including:
New Construction: Performance criteria and materials recommendations
are provided for concrete to be used in new concrete pavement
construction.
Patch or Repair: Additional recommendations specific of patch or repair
concrete mixtures are provided.
Overlay: Additional recommendations specific to concrete mixtures used
for concrete overlays are provided.
Project importance:
o High: applies to high volume traffic or Interstate highways. For this
level in-depth testing with the most sophisticated laboratory
equipment and testing procedures is anticipated.
o Medium: corresponds to medium volume traffic or State highways.
For this level, less sophisticated laboratory equipment and testing
procedures than those used for high importance are recommended.
o Low: applies for low volume traffic/ low priority roads with limited
testing budget.
Steel Reinforced Pavement: Applies for presence of any steel in the
pavement: This conditions triggers the need to look at designing concrete
resistant to corrosion.
Construction
o Early Opening: Refers to the need to open the concrete pavement
to traffic soon after construction. This condition triggers the need to
look at concrete mixtures that develop higher early-age strengths.
A drop down for the Early Opening time provides recommendations
on typical mixtures used for different early-opening times. This last
option is not currently enabled in this version of the software.
o Placement Method

Slip-form: When using slip-formed construction, a low slump
concrete mixture is recommended to prevent pavement edge
slump.

Fixed forms: A slump of 25-100 mm is recommended for
concrete mixtures where fixed forms are used.

4
 DRAFT

Figure 2: Job Specific Information Input Screen

2.3 Environmental Criteria

After entering the pavement design information, the environmental factors are
considered. There are three input screens under environment: location, climate and
exposure.

2.3.1. Location
In the Location input screen, the user is able to assess local weather by pin-
pointing the location of paving (Figure 3). Location of the project allows extracting
historical climatic data for the site where the concrete mixture will be constructed. The
climatic weather data is taken from the National Oceanic and Atmospheric
Administration (NOAA) weather database. When the project location is selected in
COMPASS, the environmental database is queried for weather data from weather
stations near the specified location. Using an intelligent algorithm, the most relevant
weather stations are selected, and the weather information for the specified location is
calculated based on a weighted interpolation scheme. The weather database contains
mean hourly readings for the entire year (based on 30 years of historical data) for a
number of weather stations distributed nationwide. In other words, the weather data is
not current or forecast but historical and the reported information is based on averages
of historical data over 30 years for the selected location.

5
 DRAFT

The user can use the navigational arrows and zoom control along the top left
corner of the map to locate the project. Once the project site is visually identified, the
user can click on the specific location in the map with the mouse. This allows the
program to pin-point the location of the project and to attach local weather patterns to
the analysis.

Alternatively, the user can enter the latitude and longitude of the paving project
under the Selected Location section. The user must click Set Location to lock in that
latitude and longitude.

Next to the latitude and longitude boxes, COMPASS shows a dialogue box for
the location and weight (%) for the three weather stations that are closest to the paving
project. In order to gain accurate weather data, COMPASS considers the three closest
weather stations to the paving project and shows the weighted average distribution of
those weather stations.

Figure 3: Location Input Screen

2.3.2. Climatic Considerations

Once the Location input screen is completed, the user can move to the Climate
input screen. The user is able to select the dates of construction and the time of day
6
 DRAFT

that the paving will take place. With this information, the historical climatic database is
queried for the climatic conditions during paving and the following climatic information is
evaluated:
o Temperature (Low and High)
o Wind Speed
o Humidity

The user may fine tune the weather data and specifically narrow down the local
weather patterns for the construction time. If there is more accurate temperature data,
the user is able to adjust those values. The wind speed and humidity (%) is listed
alongside the temperature ranges and the user is able to change those values either by
entering the values in the input boxes or by dragging the minimum and maximum
temperatures in the slider control.

2.3.3. Exposure
After considering the local weather patterns, the user is able to assess the type
of exposure the pavement will be subject to, such as an abrasive environment or
freeze/thaw environments. There are also considerations for sulfate exposure in coastal
regions. The user is able to select the box and click to select which type of exposure is
being considered. For sulfate exposure, there is a drop down menu in order to assess
the amount of exposure, from negligible to very severe.

Abrasive Environment: This indicates whether the pavement is subjected

to an abrasive environment such as heavy traffic, use of studded tires, or
a project where the concrete mixture is required to possess abrasion
resistance characteristics.
Freezing and Thawing: Corresponds to concrete exposed to freezing and
thawing cycles in a moist condition or concrete exposed to deicing salts.
Salt water, brackish water or seawater: Corresponds to concrete exposed
to chlorides from deicing salts, salt water, brackish water, seawater, or
spray from these sources that require concrete resistant to corrosion when
embedded steel is present.
Sulfate exposure: Recommendations on sulfate resistance of the concrete
mixture and recommended cement types are provided based on the level
of sulfate exposure. The sulfate exposure levels are defined according to
Portland Cement Association (PCA). 1
o Negligible: corresponds to less than 0.10% by mass of water-
soluble S04 in soil or less than 150 ppm S04 in water.
o Moderate: corresponds to 0.10% to 0.20% by mass of water-
soluble S04 in soil or 150 to 1500 ppm S04 in water.
o Severe: corresponds to 0.20% to 2.0% by mass of water-soluble
S04 in soil or 1500 to 10000 ppm S04 in water.
o Very Severe: corresponds to more than 2.0% by mass of water-
soluble S04 in soil or more than 10000 ppm S04 in water.

7
 DRAFT

2.4 Analysis
In using the COMPASS software, the user provides key information that identifies
important criteria that the user should consider based on site-specific conditions such as
structural design, environmental conditions, and construction constraints. The result
from the performance criteria analysis is a list of recommended criteria that the user
should consider when selecting materials and proportions to achieve an optimum
mixture that meet these criteria.

2.4.1. Performance Criteria Analysis

After entering all the project background data and establishing the weather data,
the user is now ready to consider what characteristics are specific to their paving
project. The list of performance criteria can be seen across the top of the Performance
Criteria screen.

The output screen for the performance criteria analysis is shown in Figure 4. The
relevant PCC properties identified appear as tabs across the top of the screen. Clicking
on the tab of interest accesses the property specific information: suggested test
methods for the projects level of importance, general construction methods to be aware
of, and suggested materials and proportions. When clicking on Read More a link to the
specific PCC property section in the FHWAs Integrated Materials and Construction
Practices for Concrete Pavement is provided for the user to research more detailed
information on the selected property.

Figure 4: Performance Criteria Analysis window

8
 DRAFT

2.4.2. Materials Selection Analysis Figure

The second analysis stage of this module refers to the Materials Selection
Analysis. In this analysis, inputs for selected material constituents in the Available
Materials screen (Figure 5) trigger Knowledge Base recommendations and warnings
based on ways to optimize the identified, relevant mixture performance criteria.

Available materials inputs that the user can select include:

Water Type
Cement Type
Fine Aggregate Type
Coarse Aggregate
o Geological Type
o Shape
o Elongated Particles
o Texture
o Maximum Size
o Gradation
o Los Angeles Abrasion
o Absorption
o Marginal Aggregate
o Recycled Aggregate
o Potential Impurities
Admixtures
o Air Entraining Admixtures
o Chemical Admixtures
o Fiber Type
Supplementary Cementitious Materials (SCM)
o Fly Ash Class
o GGBF Slag or Slag Cement
o Metakaolin: Not enabled in this version
o Natural Pozzolans
o Silica Fume: Not enabled in this version

9
 DRAFT

Figure 5: Available Materials Input Screen

Figure 6 shows the output window of this module. On this analysis window,
icons of material constituents line the top of the screen. By clicking on the icon of
interest, information on recommended materials as well as warnings on selected
materials that may compromise a given PCC property are displayed. The user can see
additional information on any given entry by clicking on the Show Details button
including the information source.

10
 DRAFT

Figure 6: Performance Criteria Output window

3. GRADATION MODULE
Figure 7 shows the Aggregate Gradation Module. The gradation module is used
to evaluate properties such as aggregate packing density and workability as a function
of user-defined or model recommended aggregate blends. When first starting the
gradation module, a general information input screen is shown first where the user can
enter the details about the project and specific notes about the analysis being
performed.

3.1 Entering Aggregate Blends

Under the Aggregates input screen, the user enters information on each of the
aggregates available.

This screen presents the available aggregates on the top half of the screen and
the aggregate general information, physical properties and sieve analysis (gradation) on
the bottom half of the screen under each of the respective tabs. The inputs entered
under each of these tabs correspond to the aggregate selected on the top half of the
screen.

The Add, Copy and Remove buttons on the top right corner are used to add new
aggregates, create a copy the inputs of an existing aggregate into a new aggregate, and
remove available aggregates from the Gradation project file. The Copy and Remove
buttons are applied to the highlighted aggregate in the list of aggregates.

11
 DRAFT

Figure 7: Aggregate Gradation Module Input Screen

Once an aggregate is added its properties need to be established. Select that

aggregate by clicking on its name and it will become highlighted. From here, the
General Information tab in the middle of the screen is used to re-name and identify the
general information about that aggregate. Under the Physical Properties tab, the user
can identify values for specific gravity, compacted unit weight, and other parameters
explicit to that aggregate. A brief description of each of these tabs follows:

General Information includes:

Aggregate name (Type a descriptive identification of the aggregate)
Aggregate source (Type in the name of aggregate source or quarry)
aggregate type
o Fine: typically for aggregates with maximum size of aggregate in
the No. 4 sieve below,
o Coarse: typically for aggregates with sieve sizes between 1
to No. 8, and
o Intermediate: Typically for aggregates with sieve sizes between
to No. 16)

12
 DRAFT

Physical Properties include:

Aggregate shape: subjective assessment of the aggregate shape (e.g.
round, semiangular, angular).
Specific gravity: Obtained through AASHTO T084 (ASTM C 128) for fine
aggregate or AASHTO T085 (ASTM C 127) for intermediate and coarse
aggregate.
Compacted (rodded) and Loose Unit weight (density): Obtained trough
AASHTO T19 (ASTM C 29).
Packing Characteristics
o Packing Density: Used in De Larrard compressible packing model
and Toufar model. May be determined as the compacted unit
weight divided by the product of the specific gravity and water
density. Not necessary for user-defined blends.
o Voids Ratio: Used in Dewar model. May be determined as the
product of the specific gravity and water density divided by the
loose unit weight. Not necessary for user-defined blends.

Gradation: Under the Gradation tab (Figure 8), the user may enter or modify the
sieve analysis for each selected aggregate, which will be used to create a
combined gradation later. To set the gradation specific to each aggregate, it
must be highlighted by clicking on the name in the list above. The user can
select to use percent passing or percent retained for analysis. Depending on
which is selected, the user can enter the laboratory analysis in the respective
column. The program provides a default set of sieves, but the user has the
option of entering additional sieve sizes in the Add Sieve section. Selected
sieve sizes can be removed with the Remove selected sieve button. NOTE:
The sieve must be highlighted before clicking remove. If an accidental deletion
of a sieve size occurs, simply click default sizes and COMPASS will restore the
original sieve sizes. By clicking on the Revert to default sieves any user-defined
sieves are deleted and the software reverts to the default set of sieves. The
graph at the bottom of this screen shows the percent passing and retained under
each sieve size for the selected aggregate on the top half of the screen.

This method is repeated for each aggregate individually, until all are entered.

13
 DRAFT

Figure 8: Aggregate Gradation Input Screen

3.2 Analysis
Before proceeding to the analysis window, select the aggregates to use in the
analysis by clicking in the box next to each of them in the list of aggregates in the
Aggregates input screen. A green check mark should appear for each aggregate
selected. This tells COMPASS which aggregates to include into the combined
gradation. After checking the desired boxes, proceed to the analysis input screen
(Figure 9). The user is then able to perform aggregate blending for multiple aggregates.

14
 DRAFT

Figure 9: Gradation Analysis window

In the analysis screen, there are two ways of blending the selected aggregate:
1. Using Custom Proportions: The user checks the Use Custom Proportions
check box and enters the proportions for each aggregate in the Custom
column under the Proportions as Percentage of Mass Table.
Alternatively for custom proportions, a control appears to select the
custom blend when blending two or three aggregates.
a. When two aggregates are blended, a sliding control appears to
select the custom blend. The blend is selected by dragging the
slider between 0 and 100%.
b. When three aggregates are blended, a ternary blend control
appears to select the custom blend. The ternary blend control has
the property that any point selected in it yields a combination of the
three aggregates that add up to a 100%. When the point is at any
given vertex, the blend equals 100% of the aggregate represented
by that vertex and 0% of the other two aggregates. When the point
is half way between one vertex and the adjacent one, the blend
equals 50% of the aggregate represented by the first vertex and
50% of the aggregate represented by the adjacent vertex and 0%
of the third vertex. In the example in Figure 10 below, the selected

15
 DRAFT

point yields a combination of 70% aggregate 1 (X1), 10% of

aggregate 2 (X2) and 20% of aggregate 3 (X3).

X1 X1=100% X1
X1= Aggregate 1
X2= Aggregate 2
X3= Aggregate 3 X3=20%

Xi=100% 70%,10%,20%
X1=70%

X3=0% X2=0%

X2 X3 X2 X3
X2=100% X1=0% X2=10% X 3=100% X2=10%

Figure 10: Ternary Blend Control

c. For more than three aggregates, the custom blend has to be

entered in the Custom column.

2. Model: This option is used to analyze the model recommended blends.

Uncheck Use Custom Proportions and select either of Toufar, Dewar, or
De Larrard models under the Model drop down menu. The proportions
for each aggregate appears in the Recommended column under the
Proportions as Percentage of Mass Table. The model option can be
used for blending up to three aggregates.

To evaluate the user-defined or model recommended blend click Analyze. This

creates a series of five graph tabs, shown across the bottom of the screen as Void
Ratio, Packing, Coarseness Factor Chart, Percent Retained, and the 0.45 Power Chart.
Clicking on different tabs will show different graphs. Figure 11 is a screenshot of a
typical 0.45 power curve analysis window for the gradation module.

16
 DRAFT

Figure 11: 0.45 Power Curve Analysis window

If the user is not satisfied with the created gradation, they can elect to use
different custom proportions. In order to do this, check the box near the top of the
screen that reads, Use Custom Proportions, and enter the desired values in the
Custom column next to the recommend values. Then click analyze again, to re-analyze
the new proportions.

4. PROPORTIONING MODULE

This module produces mixture proportions based on ACI 211/Portland Cement

Association (PCA) proportioning guidelines to obtain preliminary mixture proportions.1,2
The user can obtain mixture proportions with a minimum number of inputs. A general
information input screen is shown first where the user can enter the details about the
project and specific notes about the mixture proportioning being performed.

4.1 Mixture Design Criteria

In this input screen, the user can define what the specifications for the mixture
are, such as slump, 28 day compressive strength, and air content.

17
 DRAFT

The user may select 28-day specified compressive strength (fcr) and COMPASS
computes the required (design) average compressive strength using the method when
data are not available to establish a standard deviation (ACI 318). Alternatively, if the
user has field strength data, the user can enter the 28-day average compressive
strength and the producers standard deviation. For this method, the following
requirements must be met:

The statistical data should essentially represent the same

materials, proportions, and concreting conditions to be used in the new
project. The data used for proportioning should also be from a concrete
with an fcr that is within 7 MPa (1000 psi) of the strength required for the
proposed work. Also, the data should represent at least 30 consecutive
tests or two groups of consecutive tests totaling at least 30 tests (one test
is the average strength of two cylinders from the same sample).

If the mixture contains entrained air, the user should check the box next to the
Entrained Air, in order for COMPASS to consider it during analysis. There is also an
option to assess an air content estimate from the amount of exposure the pavement will
experience. From there, a drop down menu allows the user to select the degree of
exposure as defined by ACI 211.1.

4.2 Materials
The materials input category the user can input the types and properties of
aggregate and cementitious materials to use in the mixture proportioning.

4.2.1. Aggregates
Under this input screen, the user is able to select the number of aggregates to be
used in the mixture proportioning from a drop down menu, located on the top of the
screen. If two aggregates are selected, COMPASS uses the ACI 211.1 method for
estimating the bulk volume of coarse aggregate increased by 10% for concrete
pavement construction.

If the user created a custom blend for the aggregates, then the Custom Blending
box must be checked in order for COMPASS to consider it during analysis. For three
and four aggregates, the custom blend check box is selected by default.

From there, the user can enter the coarse aggregate maximum size, along with
other aggregate parameters, such as, coarse aggregate oven dry unit weight (to
estimate coarse aggregate content using ACI 211.1 method) and fine aggregate
fineness modulus.

There is also a table shown to the user that lists the specific gravity, moisture
content (%) and absorption (%) of each aggregate. These values can be adjusted by
clicking on the desired box within the table and entering the required value. The name
of each aggregate can also be edited here.

18
 DRAFT

4.2.2. Cementitious
After the aggregate properties are entered, the cementitious properties must be
considered. These values are entered under the cementitious input screen. The cement
specific gravity is entered. If there is a SCM being used in the mixture, the SCM content
be entered. There is a drop-down menu provided with options to change the SCM
content units from percent of cementitious by volume to percent by weight, or they can
be defined in English or SI units. For COMPASS to consider the SCM, the user must
enable that SCM check box next to the Enabled label. From there, the user can identify
the SCM name, content and specific gravity. If there is a second SCM being used, the
program provides a second area to enter the second SCM name and properties.

4.3 Analysis
After the materials are entered, it is time for COMPASS to compute the mixture
proportions.

4.3.1. Initial Mix Proportions

In this analysis screen, COMPASS displays the initial mixture proportions it
created, based on user inputs. These values cannot be directly changed, they are only
shown for the user to review. The user can make adjustments under the following two
input screens.

4.3.2. Water Adjustments

Under the water adjustments screen, the user can assign factors for water
demand adjustment, based on aggregate angularity and gradation, admixtures, SCMs,
and other factors, such as environmental. Rough water demand adjustment guidelines
are given based on These values can be adjusted by using the slider bars, shown on
the right of the screen. Under the admixtures list, the user can select which type of
water reducer (normal, mid-range, or high range) is being used by checking the box
next to the name. Once the water adjustments are finalized, the suggested maximum
water demand adjustment (%) appears at the bottom of the screen. A maximum
adjustment of 30% is recommended. However, the user can elect to use the cumulative
adjustment instead of the suggested maximum adjustment.

4.3.3. Refined Mix Proportions

After completing the water demand adjustments, the refined mix proportions are
shown under the respective tab. Figure 12 is a screenshot of the refined Mix
Proportions Analysis window. From this screen, additional mixture adjustments can still
be made. On the right hand side, there is a box that is titled Account for Moisture. This
box must be checked for COMPASS to compute the aggregate moisture and absorption
corrections based on the aggregate moisture and absorption inputs. The moisture
corrected proportions are displayed when this box is checked.

The user can also change the batch size. Additional mix adjustments can be
made to account for a maximum or minimum water to cementitious ratio, and cement

19
 DRAFT

content. The type of adjustment can be selected from a drop-down menu that includes
options for water to cementitious ratio, cement or none. If a cement adjustment is
selected, the box for cement will highlight and the corrected value can be entered. After
making adjustments, the user will see the new mix proportions. If they do not like the
new adjustments and wish to start over, clicking the reset button will restore the initial
values.

Figure 12: Proportioning- Refined Mix Proportions window

In the middle of the screen is a table showing the recommended mix values from
COMPASS. The table shows the mix proportions in four columns, each having a
different way to measure the mixture ingredients, such as by mass (lb/yd3), by volume
(ft3/yd3) or by percent of mass or volume. The values in this table cannot be directly
changed, but they change when adjustments are made to the mix design from the upper
right hand corner of the screen. There are graphical representations of the mix
proportions located at the bottom of the screen. These show the mix proportions by
percent mass and percent volume.

20
 DRAFT

5. OPTIMIZATION
The purpose of the Optimization module alone is to guide the user on identifying
concrete mixture proportions that best meet the job-specific mixture performance
criteria.

A general information input screen is shown first where the user can enter the
details about the project and specific notes about the mixture optimization being
performed.

The Optimization module is divided into two stages:

1. Trial batches and
2. Mixture optimization

Having gone through the Mix Expert, Aggregate Gradation, and Proportioning
modules, the user has a better understanding of what material constituents to focus on
and what properties to optimize for. Constituents and constituent ranges are selected
based on:
Recommendations from Mix Expert,
Recommendations from Aggregate Gradation and Mixture Proportioning
modules, and
Pertinent concrete specifications.

5.1 Trial batches

In the first analysis stage, COMPASS recommends trial batches using
experimental design techniques.

5.1.1. Materials
In the Materials input screen the user enters the materials constituents and
proportioning ranges to use in the optimization (Figure 13). The proportions of up to
five PCC material constituent factors (including water/cement or water/cementitious
ratio) can be optimized. The user can also enter additional materials with fixed
proportions.

To add a new constituent not listed, the user can click on the New button
(located below the constituent list) and create a constituent. A dialogue box will appear
and the user can define the new constituent there. The user can also remove any
constituent by clicking to highlight that constituent and then clicking the Remove button
(located below the constituent list).

By checking or un-checking the boxes next to the material constituents, the user
can select which factors to include in the optimization analysis.

To change the range of values and properties given for a constituent, the user
must click on the constituent and highlight it. Then using the Range Values and
Properties input areas below the list, the user can define the range of values or create a
fixed value that will not change.

21
 DRAFT

In the Range Values input area the user can select whether a constituent will
have a fixed amount by checking on the Fix Factor check box or specify a range of
values by leaving Fix Factor unchecked.

Figure 13: Materials Input Screen

The corresponding unit will show for the selected constituent. Aggregate and
Fiber constituent ranges can be entered as % by mass or % by volume. Admixture
constituents can be entered in different units of volume per mass unit of cement (e.g. FL
oz per 100 lbs of cement). Supplementary cementitious materials can be entered as a
% cement mass replacement.

Beneath the Range Values input area, there is a Properties area where the user
must identify the properties of the material constituent being used including material
type, specific gravity and material cost. This cost can be utilized later for optimizing the
mix.

5.1.2. Trial Batches - Analysis

Once all the material inputs are entered, the trial batches analysis screen is the
next step. In order to see the trial batches, the user must click the Create Batches
button on the top of the screen. Before clicking to create batches, the user must go back

22
 DRAFT

to the materials input screen and check to see that the desired constituents have a
check mark by their names. If there is no check mark, COMPASS will not consider that
constituent in the analysis.

The result from this analysis stage is a list of trial batches in both volumetric and
gravimetric format for the evaluation and determination of optimum mixture proportions.
The number of trial batches required is determined by COMPASS with the use of
experimental design techniques and is a function of the number of variable mixture
constituents to be optimized. Material constituents defined as fixed factors do not affect
the number of trial batches. This can be seen in Figure 14.

Figure 14: Trial Batch Analysis window

5.2 Mixture optimization

In the mixture optimization stage the user selects the mixture design criteria to
optimize for, enters the batching results, specifies desirability functions for each criterion
and performs the optimization analysis.

23
 DRAFT

5.2.1. Mix Design Criteria

Mixture design criteria of interest are required for the optimization analysis. The
design criteria are selected based on recommendations from Mix Expert, or directly
entered by the user based on constraints specified by the client.

Under the Mix Design Criteria input screen, the user can define what criteria are
important, such as cost, shrinkage, strength, etc. These criteria are shown at the top of
the screen. In order to select the criteria for analysis, the user must check the box next
to the criteria.

If there are criteria not listed that the user wishes to use, the criteria can be
created by using the New button, located under the list. This will open a new dialogue
box and prompt the user to define the new criteria. If there are additional criteria listed
that are not being used, the user can highlight them and click the Remove button to
eliminate them. The user can elect which source to use, by selecting the drop down
menu under the criteria list and choosing virtual batching, lab testing, or both. Each
source is explained below:
Lab Testing: If Lab Testing is selected as the source, the user will have to
enter the lab testing results in the Batching input screen.
Virtual Batching: Default models in COMPASS are used to predict the
batching results. This option is available only for strength, cost, shrinkage
and slump. Custom criteria must be determined through laboratory
testing.
Lab Testing and Virtual Batching: This option is only available for strength
criterion. With this option, the user may select any of the trial batches
identified in the Trial Batch analysis and enter the strength results for that
trial batch. With this information, COMPASS calibrates the default
strength models to predict the strength results for all other trial batches.

It is recommended that Virtual Batching and Lab Testing and Virtual Batching
options be used only for preliminary analysis.

Some mix design criteria have specific inputs that need to be defined by the user.
For example, for strength, the user needs to define the strength type (compressive,
flexural, or tensile) and age in days.

5.2.2. Batching Results

After defining the mix design criteria, then the user selects the batching input
screen to enter the batching results. The batching input screen is used to enter the
results from laboratory testing and/or to display results from virtual batching. In order to
view the batches, the user must click the Run button, located on the top left of the
screen. The user then enters the batching results for the criteria that have Lab Testing
selected as source in the mix design criteria screen. COMPASS shows the predicted
values for the criteria with Virtual Batching or Lab Testing and Virtual Batching as
source.

24
 DRAFT

A graphical representation of the data is shown across the bottom of the screen
to evaluate the reasonableness of the batching results. In order to change the graph to
different criteria, click on the headings in the columns in the table above. This shows
the different graphs for the different criteria. If there are more batches created that what
can be seen on one screen, place the arrow of the mouse over the table and simply
scroll through the table or place the mouse between the table and graph until a double
sided arrow appears. Click and drag the line down to show more of the table or less of
the graph.

5.2.3. Desirability
Under desirability inputs, the user identifies the value and desirability function of
each selected criterion to compute an overall desirability of the mixture. The overall
desirability for each individual mixture is computed using the desirability functions and
importance (desirability weightings) defined for each mixture design criterion.

In the Desirability input screen, there are tabs across the top of the screen that
reflects the criteria previously selected. The user can click on each tab to view the
graphical desirability of each criterion. Under each tab, the user can assign a ranking of
importance to each criterion, from 1 to 5. The user can also select how many points to
show on the graph, in the range of 3 to 10.

As Figure 15 shows, desirability functions can be in general of three different

forms depending on the criterion selected:
Target: A target desirability function is used to seek a target value such as a
target slump of 2 inches.
Maximum: A maximum desirability function is used to seek a maximum value
such as the maximum strength.
Minimum: A minimum desirability function is used to seek a minimum value
such as the minimum cost.

Desirability Target Desirability Maximum Desirability Minimum

1 1 1

0 0 0
PCC Property PCC Property PCC Property

Figure 15: Typical desirability functions

The current version of COMPASS provides ample flexibility to the user to enter
practically any custom desirability function.

25
 DRAFT

In the desirability input screen, the user modifies the desirability curve by
assigning a desirability from 0 to 100 % to each criterion value entered.

For example, in Figure 16, a maximum desirability function is entered for 28-day
compressive strength. Four points are selected to define the desirability function
corresponding to 0, 4000, 5000 and 8000 psi. For each point, a desirability of 0%, 0%,
100%, and 100 % is respectively assigned. With this inputs, COMPASS assigns a
desirability of 0% to any mixture that achieves a strength below 4000 psi. A desirability
of 100% is assigned to any mixture with a strength above 5000 psi. COMPASS will
assign an intermediate desirability for mixtures with a strength between 4000 and 5000
psi.

Figure 16: Desirability Function for Strength

5.2.4. Mix Optimization - Analysis

With the information on the batching results and assigned desirability, the
optimization analysis is run. The optimization analysis consists in generating response
surface models to predict PCC properties of interest for an extensive set of mixtures
within the experimental space defined by the constituent limits.

The optimization analysis includes various options for evaluation of the optimum
mixes. These include:

26
 DRAFT

Options to report top ten optimum mixes in volumetric and gravimetric formats,
Materials constituent factors and predicted responses.
Model-fit statistics and plots that allow the user to evaluate the goodness of fit of
the optimization results.

In order to see the results, the user must click the Optimize button. This button
is located in the upper left hand corner of the analysis screen (Figure 17). From here,
the user can see the top ten optimum mixes identified by COMPASS as a function of
the desirability functions set previously.

Under the optimization button, there is a list of 5 tabs that are used to view the
optimization process, such as gravimetric, volumetric, factors, responses and model fit.
The user can rank the values from highest to lowest by clicking on the column heading
and a grey arrow will appear. This ranks the numbers in that particular column and the
corresponding columns readjust.

Figure 17: Optimization Analysis Window

Figure 18 shows the Model Fit analysis screen. In this screen, for each criterion,
COMPASS compares the actual values determined through lab testing (or virtual
batching) with those predicted by the response surface models developed by

27
 DRAFT

COMPASS. Each criterion can be evaluated by selecting it in the Criterion drop down
menu at the top of the tab. The model r-square (r2) and standard deviation are
presented along with a tabular and graphical comparison of measured versus predicted
values. The red line in the comparison plot represents the equity line between
measured and predicted.

Figure 18: Model Fit Analysis Screen

REFERENCES

1. Kosmatka, S., Kerkhoff, B. and Panarese, W. C. Design and Control of Concrete Mixtures:
Engineering Bulletin 001. Portland Cement Association. 14th Edition. Skokie, IL. 2002.

2. ACI. Standard Practice for Selecting Proportions for Normal, Heavyweight, and Mass Concrete, ACI
211.1-91 (Reapproved 1997), American Concrete Institute, Farmington Hills, MI, 1991.

28