Research Papers

Bayesian Inference for Milling Stability Using a Random Walk Approach

[+] Author and Article Information
Jaydeep Karandikar

Georgia Institute of Technology,
Atlanta, GA 30332
e-mail: jaydeep.karandikar@me.gatech.edu

Michael Traverso

Stanford University,
Stanford, CA 94305
e-mail: miketrav@stanford.edu

Ali Abbas

University of Illinois at Urbana-Champaign,
Urbana, IL 61801
e-mail: aliabbas@illinois.edu

Tony Schmitz

University of North Carolina at Charlotte,
Charlotte, NC 28223
e-mail: tony.schmitz@uncc.edu

Contributed by the Manufacturing Engineering Division of ASME for publication in the JOURNAL OF MANUFACTURING SCIENCE AND ENGINEERING. Manuscript received July 29, 2011; final manuscript received February 17, 2014; published online April 11, 2014. Assoc. Editor: Suhas Joshi.

J. Manuf. Sci. Eng 136(3), 031015 (Apr 11, 2014) (11 pages) Paper No: MANU-11-1258; doi: 10.1115/1.4027226 History: Received July 29, 2011; Revised February 17, 2014

Unstable cutting conditions limit the profitability in milling. While analytical and numerical approaches for estimating the limiting axial depth of cut as a function of spindle speed are available, they are generally deterministic in nature. Because uncertainty inherently exists, a Bayesian approach that uses a random walk strategy for establishing a stability model is implemented in this work. The stability boundary is modeled using random walks. The probability of the random walk being the true stability limit is then updated using experimental results. The stability test points are identified using a value of information method. Bayesian inference offers several advantages including the incorporation of uncertainty in the model using a probability distribution (rather than deterministic value), updating the probability distribution using new experimental results, and selecting the experiments such that the expected value added by performing the experiment is maximized. Validation of the Bayesian approach is presented. The experimental results show a convergence to the optimum machining parameters for milling a pocket without prior knowledge of the system dynamics.

Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.


Tlusty, J., and Polacek, M., 1963, “The Stability of Machine Tools Against Self-Excited Vibrations in Machining,” ASME Int. Res. Prod. Eng., 1, pp. 465–474.
Tobias, S. A., and Fishwick, W., 1958, Theory of Regenerative Machine Tool Chatter, The Engineer, London, 205, p. 258.
Merrit, H., 1965, “Theory of Self-Excited Machine Tool Chatter,” ASME J. Eng. Ind., 87(4), pp. 447–454. [CrossRef]
Arnold, R., 1946, “The Mechanism of Tool Vibration in the Cutting of Steel,” Proc. Inst. Mech. Eng., 54, pp. 261–284. [CrossRef]
Komanduri, R., 1993, “Machining and Grinding: A Historical Review of the Classical Papers,” ASME Appl. Mech. Rev., 46(3), pp. 80–132. [CrossRef]
Tlusty, J., 1978, “Analysis of the State of Research in Cutting Dynamics,” Ann. CIRP, 27(2), pp. 583–589.
Smith, S., and Tlusty, J., 1991, “An Overview of Modeling and Simulation of the Milling Process,” ASME J. Eng. Ind., 113(2), pp. 169–175. [CrossRef]
Smith, S., and Tlusty, J., 1993, “Efficient Simulation Programs for Chatter in Milling,” Ann. CIRP, 42(1), pp. 463–466. [CrossRef]
Ehmann, K., Kapoor, S., DeVor, R., and Lazoglu, I., 1997, “Machining Process Modeling: A Review,” ASME J. Manuf. Sci. Eng., 119(4B), pp. 655–663. [CrossRef]
Smith, S., and Tlusty, J., 1997, “Current Trends in High-Speed Machining,” ASME J. Manuf. Sci. Eng., 119(4B), pp. 664–666. [CrossRef]
Merchant, M. E., 1998, “An Interpretive Look at 20th Century Research on Modeling of Machining,” Mach. Sci. Technol., 2(2), pp. 157–163. [CrossRef]
Altintas, Y., and Weck, M., 2004, “Chatter Stability of Metal Cutting and Grinding,” Ann. CIRP, 53(2), pp. 619–642. [CrossRef]
Kim, H. S., and Schmitz, T., 2007, “Bivariate Uncertainty Analysis for Impact Testing,” Meas. Sci. Technol., 18, pp. 3565–3571. [CrossRef]
Schmitz, T., and Duncan, G. S., 2005, “Three-Component Receptance Coupling Substructure Analysis for Tool Point Dynamics Prediction,” ASME J. Manuf. Sci. Eng., 127(4), pp. 781–790. [CrossRef]
Duncan, G. S., Kurdi, M., Schmitz, T., and Snyder, J., 2006, “Uncertainty Propagation for Selected Analytical Milling Stability Limit Analyses,” Trans. NAMRI/SME, 34, pp. 17–24.
Mann, B. P., Bayly, P. V., Davies, M. A., and Halley, J. E., 2004, “Limit Cycles, Bifurcations, and Accuracy of the Milling Process,” J. Sound Vib., 277(2), pp. 31–48. [CrossRef]
Howard, R., 1966, “Information Value Theory,” IEEE Trans. Syst. Sci. Cybern., 2(1), pp. 22–26. [CrossRef]
Howard, R., 1970, “Decision Analysis: Perspectives on Inference, Decision and Experimentation,” Proc. IEEE, 58(5), pp. 632–643. [CrossRef]
Schmitz, T., and Smith, K., 2009, Machining Dynamics: Frequency Response to Improved Productivity, Springer, New York.


Grahic Jump Location
Fig. 9

Posterior cdf for milling stability given a stable test result (left) and an unstable test result (right) at an axial depth of 5 mm and spindle speed of 7000 rpm

Grahic Jump Location
Fig. 8

Updated cdf at 7000 rpm given a stable test result (left) and an unstable test result (right) at a test axial depth of 5 mm

Grahic Jump Location
Fig. 7

Sample paths remaining after filtering given a stable test result (left) and an unstable test result (right) at and axial depth of 5 mm and spindle speed of 7000 rpm

Grahic Jump Location
Fig. 6

Prior probability of stability in the spindle speed-axial depth domain (left). The probability of stability is 0 at an axial depth of 10 mm.

Grahic Jump Location
Fig. 5

Histograms of axial depths at 4000 rpm (left) and 10,000 rpm (right)

Grahic Jump Location
Fig 4

10,000 sample paths after filtering. The paths that cross 0 or 10 mm in the spindle speed range of 4000 rpm to 10,000 rpm have been removed.

Grahic Jump Location
Fig. 3

Many sample paths generated in the spindle speed-axial depth domain

Grahic Jump Location
Fig. 2

Histograms of x at 5 s (left) and 10 s (right)

Grahic Jump Location
Fig. 1

Twenty (left) and 5000 (right) random walks with a normally distributed position step size described by N(0,0.1)

Grahic Jump Location
Fig. 18

Test point selections compared with the analytical stability lobes

Grahic Jump Location
Fig. 19

Experimental validation of the stability lobes

Grahic Jump Location
Fig. 10

Tool path for pocket milling

Grahic Jump Location
Fig. 11

Cost of machining at axial depth–spindle speed combinations given that the resultant cut is stable. Notice the steps in the cost function at integer fractions of the pocket depth.

Grahic Jump Location
Fig. 12

Stability results for the value of information testing. The “o” symbols represent a stable result and the “x” symbols indicate an unstable result.

Grahic Jump Location
Fig. 13

Experimental setup for stability testing

Grahic Jump Location
Fig. 14

Frequency content of the acceleration signal (left) and the machined surface (right) at {10,000 rpm, 6.25 mm}. Content is seen only at the tooth passing frequency (167 Hz) and its harmonics.

Grahic Jump Location
Fig. 15

Frequency content of the acceleration signal (left) and the machined surface (right) at {8294 rpm, 8.34 mm}. This unstable cut exhibits content other than tooth passing frequency and its harmonics (left) and chatter marks are observed (right).

Grahic Jump Location
Fig. 16

Posterior stability cdf after 20 tests

Grahic Jump Location
Fig. 17

FRFs for the flexure in the x (left) and the y (right) directions used in the experiments. Note that the dynamic stiffness is an order of magnitude higher in the y direction.

Grahic Jump Location
Fig. 20

The updated cdf at 6500 rpm given a stable test at {7000 rpm, 5 mm} with standard deviations of 0.5 mm and 0.25 mm for the random walk generation

Grahic Jump Location
Fig. 21

Updated posterior cdf given a stable test result at {7000 rpm, 5 mm} for random walks generated using standard deviations of 0.5 mm (left) and 0.25 mm (right)

Grahic Jump Location
Fig. 22

Updated posterior cdf after 20 tests using random walks generated with standard deviations of 0.5 mm (left) and 0.25 mm (right)

Grahic Jump Location
Fig. 23

Test point selection compared with the analytical stability lobes

Grahic Jump Location
Fig. 24

Posterior cdf of stability after eight tests at a spindle speed range of 4000 rpm to 8000 rpm



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In