Mitchell Collision Estimating Guide

[Gibert Chisholm]

Sowhat is this process called feather, prime and block? It seems that it is always a point of contention between third-party payers and the repair facility, with the biggest discussion surrounding the need and who should be doing it. I hope that throughout this article I can offer some clarification by providing facts and eliminating opinions.

I found a Collision Industry Conference (CIC) statement that says: The repair process associated with damaged painted body panels typically involves multiple operations: body repair, feather, prime, block, and refinish. The body repair process includes metal finishing and/or the use of body fillers to return the body panel to its undamaged contour. The repaired area is finished to 150-grit and is free of surface imperfections. Feather, prime and block are not-included refinish operations that complete the process from 150 grit to the condition of a new undamaged panel. The refinish process starts at the condition of a new undamaged panel and is outlined and documented in printed and/or electronic time guides. The body/paint labor and materials necessary to prepare the repaired area from 150-grit to the condition of a new undamaged part is a valid and required step in the process. The labor and material allowance for these operations requires an on-the-spot evaluation of the specific vehicle and damage.

Feather, Prime & Block is the not-included refinish operation that completes bodywork repair from 150-grit smoothness to the condition of a new undamaged panel, and the point at which refinish labor time begins. The labor and materials associated with feather, prime and block may vary depending upon the size of the repair area and should be evaluated when determining the work to be performed.

Through this research, I learned that CCC/MOTOR and Mitchell agree that feather, prime and block is in fact a refinish process. Reviewing the steps identified in the 3M SOP makes it difficult to understand how it could be included in any repair procedure. However, the amount of time required to perform the process is still not identified. During my damage appraisal documentation workshops, I encourage the attendees to determine the time by using a percentage of the panel repair time. I suggest they model their calculation similar to De-Nib & Polish calculations as shown in the CCC/MOTOR P-pages. If there were 3.5 hours of repair time on a panel, you would multiply 3.5 by the percentage you selected to reach the feather, prime & block refinish labor time. Also, by using the formula above for each repaired panel you can develop consistency in your process that will help you gain reimbursement.

As I circle back to my opening paragraph, I hope I have offered some clarification to the feather, prime & block process, why it is needed, and who should be performing the steps. Since the refinish time premise defined in the three estimating databases mentioned is based on one color applied to a new undamaged panel, it is a required refinish operation to restore a repaired panel to that new undamaged condition. With the clarification provided in this article, I feel comfortable that the information will help you overcome any contention with third party payers. The key is in the documentation and educating those opposed to the need for this required process.

John Shoemaker is a business development manager for BASF North America Automotive Refinish Division and the former owner of JSE Consulting. He began his career in the automotive repair industry in 1973. He has been a technician, vehicle maintenance manager and management system analyst while serving in the U.S. Air Force. In the civilian sector he has managed several dealership collision centers, was a dealership service director and was a consultant to management system providers as an implementation specialist. John has completed I-CAR training and holds ASE certifications in estimating and repair. Connect with Shoemaker on LinkedIn.

The RHINO software consists of a collection of modules, embedded ina distributed and asynchroneous software architecture. There are two overview papers of the RHINO system. Thefollowing and incomplete list contains some of the fundamentalmodules.

Being a sophisticated tour guide requires carefulplanning at an abstract or task-oriented level, takinginto account issues such as the amount of timeavailable, choosing appropriate exhibits to visit onthat tour, and planning the shortest path coveringthose exhibits. For that purpose we are using GOLOG, adeclarative language based on the situation calculusand developed at the University of Toronto.GOLOG, which is implemented in Eclipse Prolog, allowsthe user to concentrate on the high-level specificationof complex actions without having to worry about howthese are actually carried out by the robot. Besidesapplying GOLOG to real world scenarios, the group isalso engaged in exploring foundational issues such asthe interaction between action and knowledge andextending the expressiveness of the languagein collaboration with the University of Toronto.

H.J. Levesque, R. Reiter, Y. Lesprance, F. Lin and R. Scherl.GOLOG: A Logic Programming Language for DynamicDomains. To appear in Journal of Logic Programming,Special issue on Reasoning about Action and Change.

RHINO employs the dynamic window approach for avoidingcollisions with obstacles (such as rapidly moving humans). Thedynamic window approach is a reactive, in the sense that it considersonly a very short history of sensor readings. A key feature of thisalgorithm is that it correctly and in an elegant way incorporates thedynamics of the robot. This is done by reducing the search space tothe dynamic window, which consists of the velocities reachable withina short time interval. Within the dynamic window the approach onlyconsiders admissible velocities yielding a trajectory on which therobot is able to stop safely. The dynamic window approach can quickyreact to unexpected and rapidly moving obstacles.

D. Fox, W. Burgard, and S. Thrun. Controlling Synchro-driveRobots with the Dynamic Window Approach to CollisionAvoidance, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'96), Japan, 1996.

RHINO is capable of learning its maps, based on sensor data. Itinterprets its sensor reading using a neural network, which is trainedto convert proximity readings into circular, metric maps. Local metricmaps constructed from different readings are integrated using Bayes'srule. From the metric maps, topological graphs are constructed thatdescribe the environment in an abstract and compact way.

RHINO's motion planner is based on value iteration, a version ofdynamic programming. Value iteration propagates values (very much likean activation spreading algorithm) through the map, to find minimumcost paths to a goal. The current implementation pre-plans for everypossible exception, and also provides an optimized re-planner forefficiently adapting plans to a changing map. The planner is anany-time algorithm, i.e., it can be used to determine a robot's motionat any time, even before the value iteration process is completed.The following paper also describes novel version of motion planningusing the topological maps, which enables the robot to pre-plan everymotion plan (assuming the map is static); making motion planning aproblem that can be solved by table lookup.

S. Thrun and A. Buecken, 1996. Integrating Grid-Based and Topological Maps for Mobile Robot Navigation, National Conference on Artificial Intelligence, AAAI-96.For more detail, see also our Technical Report: Learning Maps for Indoor Mobile Robot Navigation

Localization is the problem of finding out where a robot is.Accurate localization is essential for safe operation and for reliablyperform tasks such as the one faced by RHINO. RHINO applies therecently developed position probability grid approach to solveits localization problem. The basic idea here is to estimate aposition probability density function by repeatedly matching thesensor readings with a model of the environment. This approach isparticularly well-suited for dealing with ambiguities, noise, andmodel inaccuracies in large-scale environment (50 by 100 meters orlarger). These featuress are essential for the success of this project,since they allow RHINO to know its position even in populatedenvironments such as the "German Museum Bonn".

Example of RHINO's belief state during global localization in the "German Museum Bonn". Shown in blue are is a two-dimensional map of the musem.The greyscale depicts the likelihood, assigned to each individual locationafter approximately 2 minutes of autonomous robot operation. At this point,the robot has successfully determined its location - completely by itself!

W. Burgard, D. Fox, D. Hennig, and T. Schmidt. Estimating the AbsolutePosition of a Mobile Robot Using Position ProbabilityGrids, Proceedings of the Thirteenth National Conference onArtificial Intelligence (AAAI'96).

One of the fundamental research interests of the RHINO group is robotlearning. How can we build robots that have the ability to improve bythemselves? The RHINO group believes that the ability to learn fromexperience is a crucial component of autonomous robots. Below are afew papers that investigate issues of robot learning.

3a8082e126