Page 10
www.us-tech.com
September, 2021 TECHWATCH Real Time is Too Late for Manufacturers By Dr. Michael Grant, CTO, and Nicol Ritchie, Content Writer, DataProphet
derstand, let alone resolve. Man- ufacturing equipment and pro - cess technology are advancing quickly, driving tighter toler- ances with less margin for error. In a bid to regain control
R
over the growing complexity, much emphasis has been placed on process optimization, pro- duced from a mass of real-time data. However, this data is only available as a reaction to process anomalies when they occur on the line, while the insidious root causes may remain hidden. A solution to this is a non-dis-
ruptive data-driven approach to prescriptive analytics. With prescriptive analytics,
“AI-as-a-Service” automates root cause analysis, and in some cases, can loosen a particular non-criti- cal tolerance, compensating for in- evitable upstream variance and
oot causes of systemic is- sues during production are difficult to identify and un-
ensuring holistic production line control.
FMEA Failure mode and effects
analysis (FMEA) is a typical ap- proach to troubleshooting a pro- duction issue. This method corre- sponds observed symptoms with a possible cause. But, while one fire is put out, another may flare up, requiring immediate atten- tion. The opportunity costs of stopping production are unac- ceptably high. If the rate of failure is ex-
treme, the production team must conduct a root cause analysis which, in turn, prompts the qual- ity team to produce a corrective action request indicating the need for a course correction. For example, the product engineer- ing team may then be required to generate a change note regard- ing the product’s design or process of manufacturing.
High Performance Chemistry for Advanced Technologies
This is time-consuming and
relies on narrow human expertise gained over years of experience. While it is an effective way to solve a particular issue, it rarely gets to the heart of the problem. The more a control plan is
built on narrow tolerances, the greater the chance that one of its values going out of control will create a cascade of failures. It is also important to understand the limitations of tight feedback loops to advanced manufacturing pro - cesses, which are inherently com- plex.
Like Dominos A tight feedback loop for a
production process comprises in- put (setpoint adjustments), ac- tion (a production run), feedback such as throughput versus latent defects/quality, and evaluation. However, feedback loops are
typically constrained to the dis- crete inputs of specific experts, who feed adjustments into the beginning of the next iteration of the process. Because of this, feedback loops tend to apply to singular subprocesses and are variegated according to disci- pline, team and human capacity. If we consider that such
tight feedback loops do not ac- count for the domino effect pres- ent in complex manufacturing processes, the opportunity cost to potential process optimization in this approach becomes clear. Common sense tells us that the output of each step serves as an input to the next step. Because manufacturing pro -
Pallaspeed® 990 Pure Palladium deposit for high-volume electronic applications.
• High Ductility/Low Stress • Bright Deposit • No Microcracking • Neutral pH • Wide Current Density Range • Superior Corrosion Resistance
cesses are by nature multivari- ate, the impact of any univariate deviation in one step is felt in the next, as well as within the sub- step itself. The relative impact of
process deviance is compounded when the downstream process parameters towards the end of a manufacturing sequence are hy- persensitive to (i.e. cannot toler- ate) the accumulated deviation in earlier upstream processes.
AI-as-a-Service Classically predictive artifi-
Learn more about Technic’s industry-leading plating technologies for tin, nickel, gold, silver and palladium at
www.technic.com
cial intelligence (AI) is too static to realize process optimization. For this to happen, a production system needs to be looked at holistically, the AI must be de- signed to work proactively, and it must be deployed as a service.
The optimization of a com-
plex process is guided not merely by process data about current or imminent failures, but by what the data is communicating about all the interdependent variables, including those that have histori- cally produced the best results. It continually and preemptively de- termines which variables, if ad- justed, are most likely to achieve an optimal production run. By gaining a holistic view of
any complex manufacturing process, DataProphet PRE- SCRIBE learns the relevant in- terdependencies between the many production process param- eters, including those upstream and downstream of each process. PRESCRIBE then accurately projects the impact of set-point changes to a plant’s control plan, and prescribes the next best, highest impact step towards the “best of best” (BOB) region. This approach guarantees
that suboptimal production out- comes are relegated to potential- ities, because they are solved be- fore they are realized. Results have consistently
shown that establishing a BOB region via the embedded expert guidance of data-driven machine learning provides a stable target, empowering plant controllers to sidestep the pitfalls of real-time and reactive optimization. AI in manufacturing is only
worth the investment if it proves to be a dependable tool towards achieving specific Industry 4.0 am- bitions. Its ROI must be measura- ble within a reasonable time frame. Technically, it must learn the complex interconnectivity of many process parameters, then deliver easily applicable recom- mendations for optimal plant and product metrics ahead of real time. Knowing what is happening
in real time, knowing what is likely to happen, and knowing precisely which changes to make to manifest a desired future pro- duction outcome is crucial. For this reason, a reactive paradigm based on real-time data is too late for those pursuing smart manufacturing. Contact: DataProphet, 74
Prestwich Street, 109A The Foundry Building, Green Point, 8005, Cape Town, South Africa % +27-21-300-3555 E-mail:
info@dataprophet.com Web:
www.dataprophet.com r
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100
