Challenges

CLOSED ON: 18 SEP 2019  |  REWARD: INR 3,00,000
Reward money is paid in exchange of legally acquiring the solution, implementing it to solve the problem and meeting the success criteria. Milestones for paying the reward money would depend upon the complexity of challenge and maturity of the proposed solution, which would be discussed with the solver as soon as the proposed solution is selected by us.
Closed

Short Description:

Develop an automated system for inspection of surface defects in processed steel blanks.

Challenge Details

We process hot rolled and cold rolled steel coils by slitting and cutting operations to produce them in desired sizes. The processed rectangular sheets are commonly referred to as “Blanks”. Blanks are of wide-ranging sizes, up to 1600 mm x 2500 mm in size for CR. These Blanks are used by automotive customers and white goods industries. They have stringent quality norms for strip surface quality. Thus, the Blanks need to be inspected on both top and bottom surfaces to identify the defects.

There could be various kinds of defects like Dent Marks, SPM Matte less Mark, Roll Mark, SPM Pinch Mark, Rubbing Mark, Scratch, Sliver, Lamination/ Flying scab, Gauge Mark, Carbon soot mark, Oxidation mark, Rust patch, Holes, Black/brown patches, Coils slips, Fold marks, etc. The minimum size of the defect to be detected is 5mm. Click on the following link to look at the images of the defects:

https://www.dropbox.com/s/nceizhf00pdlc4f/Visual Standard of RM defect.pdf?dl=0

While inspecting the Blanks, each Blank is inspected by the operator manually at the offline inspection stations. To perform the inspection, the operators (2 or 3) picks up the Blanks or a lifting arm (as depicted in the image) assists the operator to inspect both the sides of the blank visually. An operator takes roughly 20 secs to inspect per side of Blank.  After the inspection, the operator segregates the defective Blanks from the good ones. Though the operator is able to identify all the defects, this process of manual inspection causes operator fatigue. Moreover, this process is very time consuming, tedious and inefficient process resulting in low productivity.

Though commercial inspection systems are available in the market, the costs are prohibitively high. It is not feasible to install them in multiple processing lines either, as the volumes do not justify the high capital investment in this process.

Hence, an automated system is required to inspect these Blanks and display the results to the operator to make a decision. The system should meet the following requirements:

  1. The Inspection must be done on both the sides of the Blank.
  2. Complete automatic setup
  3. Must be able to detect defects as low as 100 microns.
  4. Alert/warn the operator in cases when defect severity goes beyond the acceptable limits.
  5. Must have the facility to identify and classify defects and present an analysis of severity & frequency of defects logged. Defects thus identified need to be registered in the form of a library. A self-learning application would be essential to enrich and make use for auto-detection of defects.
  6. The solution should be easy to integrate with the existing setup and must have a minimum footprint.
  7. Must not increase the time of inspection and should effectively increase productivity.
  8. The display to show the defects and the user interface has to be user-friendly.
  9. The system should be cost effective and easy to maintain.

Have any query or need more clarification about this challenge?