1. Sustainment Management Systems
BUILDER, PAVER, RAILER and ROOFER

2. Condition Index Development
Condition Assessments and Indexes focused on component level:
BUILDER: Roof Membrane, Air Handling Unit, Overhead Door
PAVER: Airport taxiway, apron, city block
RAILER: Turnout, Grade Crossing, Ties
ROOFER: Entire Roof Section
To make sure we’re clear, especially depending on how other presenters have presented their approaches, let’s take a moment to cover “what” our condition indexes actually focus on. The base, or foundational, Condition Assessment (and resulting Index) that each of these tools utilize is performed at the individual asset level, not the system, or facility level. This is important, as we’re more closely describing the behavior of one specific asset, and not just the overall performance of the system it is in.

3. Expertise in a Box
Condition Indexes developed by interviewing a panel of experts to get a statistical agreement on how defined defects result in a rating (0-100) of the asset’s condition.
Results in a system which incorporates experts’ experience in objectively rating asset condition using observed distresses.
Again, just a history lesson to describe how we arrived at our solution. There’s a lot of talk among facility management programs about “curves” in their system, but questioning them deeper reveals that they are typically talking about lifecycle curves, rather than condition index rating curves. This fundamental difference means they are still based upon a subjective rating and then statistically applying a trend to that subjective rating. In comparison, the SMS (or EMS) approach surveys a panel of experts in the field to get a collective agreement on the numerical rating of component given a defined scenario (of distresses). The experts were given “guidance” that put scores into ranges. This guidance effectively helped relate the two separate, not orthogonal but not aligned, issues of cost to repair and criticality. By providing bands of suggested scores, experts were able to express the differences in criticality of two different similarly-costed defects.

4. Condition Prediction
BUILDER 3.0 incorporates a new condition prediction engine based upon industry models of failure probability distribution.
Using the Weibull cumulative probability distribution function as the basis for condition trend, condition history is used to adjust the curve to the specific asset’s observed behavior.

5. Condition Prediction (cont.)
Incorporates both distress survey (traditional SMS inspection), and direct rating (quick red, amber, green rating) methods
Affects of repairs are modeled.
Patent currently applied for this method under COE-582CIP
For more information, please see “Condition and Reliability Prediction Models using the Weibull Probability Distribution” by M.N. Grussing, D.R. Uzarski, and L.R. Marrano, presented at the ASCE Applications of Advanced Technology in Transportation 2006 (AATT) Conference, August 14th-16th, 2006.

6. Work Identification - Standards
Work is automatically generated by looking for components which don’t meet a required condition level.
Standards define both minimum condition levels for work (resulting in repairs); and minimum levels of reliability (expressed in remaining service life and resulting in replacement).
Standards define the trigger point at which work is triggered (for repair). In this context serviceability is defined as the intended product of this component-section. For example, a clogged filter, or low coolant level may mean the air conditioning unit is no longer providing enough cool air (it’s service). However, this can be fixed by replacing the filter/recharging the coolant, thus restoring it’s serviceability. In this way, it’s difficult to express a direct relationship between reliability (which typically is remediated with replacement) and condition, since condition may also be affected by degradation in the service of a particular component.
Condition Index
Definition
100
Entire component-section or sample free of observable defects
99-93
No component-section or sample serviceability or reliability reduction.
92-86
Slight or no serviceability or reliability reduction overall to component-section.
85-75
Component-section serviceability or reliability is degraded but adequate.
74-65
Component-section serviceability or reliability is definitely impaired.
64-56
Component-section has significant serviceability or reliability loss.
55-37
Significant serviceability or reliability reduction in component-section.
36-11
Severe serviceability or reliability reduction, such that it is barely able to perform.
10-0
Overall degradation is total.

7. Work Identification - Policies
Policies define how standards are applied to inventory.
Policies defined upon the criticality, consequences of failure and level of risk you are willing to accept for various types of inventory (i.e. building use type, system, importance of facility, tenant requirements, etc.)
Policy sequences allow for multiple criteria on each asset, with overriding preference
Work is created by rules, based upon objective data, rather than subjectively based upon skill, knowledge, and preferences of observer
The user configures “rules” to specify which standards are applied to which component-sections. In the case multiple rules apply, a priority order is specified to identify which standard ultimately applies to the section.

8. Work Identification - Prioritization
Prioritization is used to rank work requirements
Use various parameters including economic, criticality, and geographic factors
Effectively target work to most important items by cost and criticality to mission accomplishment.
Prioritization is the process by which each work requirements is objectively ranked against each other. When we don’t have enough money to do everything, this is when the “cut” line comes in to determine what is not accomplished this year.
It’s important to stress here that we can use this decision process to compare apples to oranges by translating different metrics for different attributes which represent the same outcome (i.e. failure). We might look at one set of information when we want to consider the reliability of a roof membrane, versus the reliability of an air handling unit. But ultimately, each of these will have an expression of reliability that we can compare together to decide, given scare resources, which asset should be worked on. In the same fashion we could determine the cost of failure and/or economic benefits of each work item (again using different attributes if necessary) and then compare as an additional parameter in scoring algorithm.

9. Work Identification - Summary
Work is automatically created based upon objective rules and acceptable levels of risk (expressed as condition).
Work costs are derived automatically as a parameter of the condition.
Work is prioritized based upon user-specified criticality and financial metrics and weighted according to user business practices.
Unfunded work (based upon priority) is added to backlog, and FCI is automatically calculated.
Condition Prediction models can be applied in simulation to develop multi-year work plans for strategic and tactical planning.

10. IMPACT Integrated Multi-year Prioritization Analysis Consequences Tool
Simulation tool to allow user to vary standards, policies, prioritization, inflation, funding, uses, and real property inventory to determine effects of various facility management decisions.
Goal is to incorporate the various asset domains on the installation to establish an integrated maintenance plan that determine who to best utilize

11. Mission Focused Facility Investments
Objective: Equitably distribute facility investments to assure mission accomplishment and greatest cost effectiveness.
Method:
Utilize METL information to organize and define hierarchy of tasks.
Identify facility characteristics required to support mission
Physical Attributes (Site, size, footprint, etc)
Spaces (Changing the use of interior areas)
Components
Compare mission facility requirements against facility performance
Capability (measured by functionality)
Serviceability & Reliability (measured by condition)
Identify options for closing gaps
Track building performance and its impact on mission
MFFI is our current work research product to define the framework how all these user-defined choices will be made. METL gives us a convenient framework that already organizes the “network” nature of the organization and the expectations of what must be performed by a facility occupant. This acts as the vehicle to help the customer’s needs be communicated to the facility operator.

12. SMS Current Status BUILDER PAVER RAILER ROOFER
Used by USMC, DoE, FBI, and many private organizations.
BUILDER 3.0 is new web-based version with support for entire enterprises.
PAVER
Used by all DoD services, as well as FAA.
New desktop release - PAVER 6.0
RAILER
Used by Army, Navy, and USMC. IMA performing more implementations next year.
RAILER 6.0 is new version with redesigned user interface and support for modern and future operating systems.
ROOFER
Used by Army and Navy.
Development underway to upgrade to web-based system based upon BUILDER application framework.
Joint SMS Users Group Meeting
First meeting of multiple-program users to define common development needs and standardize on common work management practices using SMS products.

13. Questions?

14. About this Presentation
Presenter: Lance Marrano
Engineer Research & Development Center, CERL
Phone: