BLOG
BLOG
Drishti 이사회의 설립자이자 회장인 Prasad Akella박사는 기술을 사용하여 인간의 능력을 확장하는 세 번째 거대한 시장 카테고리를 만들고 있습니다. 1990년대에 프라사드는 세계 최초의 협동 로봇 (‘cobots’, 2025년까지 120억 달러의 시장이 될 것으로 예상되는 예상됨 코봇)을 개발한 제너럴 모터스 팀을 이끌었습니다. 2000년대 초, 소셜 네트워킹의 선구자인 Spoke의 공동 설립자로서, 현재 수조 달러의 가치를 가진 거대한 소셜 그래프를 구상하고 그 구축을 지원했습니다. 현재는 Drishti에서 AI 기반 생산이라는 형태로, AI의 인식력과 공장에서의 인간의 유연성을 조합하는 것에 임하고 있습니다. Prasad는 캘리포니아 마운틴 뷰에 기반을 두고 있습니다.
For a large manufacturer, everything is time-sensitive — production deadlines, project deadlines, ship dates, resolution dates, forecasts and employee schedules. The pressure to perform and hit tight timelines is high and constant. Another reality is that no matter how smoothly your production runs, you will encounter problems. Time pressure and production problems can combine to create the perfect storm. Even with the correct systems in place – PDCA and A3 report at the ready – workers and line leadership often struggle to solve problems properly in this context — if they’re solving the right problems at all.
The situation
Imagine you’re a line worker assembling hundreds of medical probes over the course of your shift. You know that your production goal is to move 300 units through your station by the end of the day. You’re on a tight schedule, which is nothing new, but this week your line is running behind schedule because one of your co-workers had a family emergency. You know the team can make up the difference but they’re going to have to work extra hard.
You notice that every once in a while one of the special caps used on your product doesn’t screw on easily. You observe that there are metal pieces in the threads suggesting that there was a machining problem. This is unusual and has never happened before, so in the name of keeping up and not causing a stir, you take a spare adapter and set of pliers and force the cap on, clearing the threads so they can go onto the product. It’s not a big deal, but it is going to add time to each cycle – ensuring the cap can screw on smoothly – making the deadline even harder to achieve. You could notify your line leader to get in touch with engineering to assess the issue, but you don’t want to be blamed if you bring the line to a halt or miss the production deadline.
It’ll be tight, but since this issue hasn’t happened before you decide that you’ll just keep on going with your temporary fix — after all, this will probably never happen again.
Luckily it works out this time. You make the production goal and keep your bosses and customers happy. No reason to bring up the issue as you don’t see any more caps like this toward the end of this production lot.
Three months later…
Everything went smoothly — thankfully no products were returned and the question of a recall was avoided, especially given the risky nature of med device manufacturing. But the next shipment of specialized caps comes in and the same issue occurs from months before. Once again it puts production goals in jeopardy and once again the technician elects to use their workaround.
The point
These issues happen all the time, even with the best intentions from your workers. In the ideal situation, the andon cord should be pulled, the issue reported and solved immediately production will slow until the problem with the cap was resolved and risks mitigated. However, in the real world, we know that there is a gray area where production goals are prioritized and issues are not flagged.
At the very least, the problem should be recorded. In the case of the cap – assuming the cap was an externally sourced part – a complaint and perhaps a supplier corrective action (SCAR) should have been filed with the provider.
What we can all agree is that non-trivial issues like that cited in the example are not for the line associate to assess and solve for themselves.
The solution
To mitigate these kinds of in-line problems that occur during the middle of busy production cycles we need to set up a system by which problems are flagged, documented and available for viewing by those who have the knowledge to fix the issue. In lean manufacturing, PDCA and A3 reports often serve as a way to capture and solve these issues. Even if a work-around solution is needed in the moment, it is a best practice for quality and/or process engineering to document what the change was and whether or not there was any risk to the final product.
The need for these systems seems obvious and is in place in most major manufacturers but it at odds with the realities of time pressures and trust from line workers. Drishti recently conducted a survey surrounding the perceptions of manufacturing from 500 non-management workers. Of the survey respondents with manufacturing experience, only 28% believed that factory workers are treated with respect and trusted by management. This is a telling figure.
So how do we get line workers to participate in the problem solving process, the correct way, every time, as issues occur? We believe it starts with empowering the workers and building trust.
Drishti provides exactly the kind of vigilance, real-time insight and collaboration that can combat these situations as they occur. Using our AI-powered platform our customers (some of the biggest and most lean manufacturers in the world) can see when cycle times are over the expected amount and whether or not standardized work instructions are followed in the right order and/or correctly performed. Think of Drishti in part as an electronic, automatic andon that consistently spots and reports deviations from the standard.
The recorded video is tagged and searchable so that the information about a particular event is never lost in the shuffle again. It is the ultimate warning system with the capability of sending messages to whoever needs the information. More problems will be solved correctly, risks will be more apparent and your workers will feel empowered knowing that they won’t be blamed for issues that are not their fault. Ultimately, production and quality will increase, adding to organizational health and peace of mind for all involved.
Interested in learning more about Drishti’s solution? Learn how Drishti's patented action recognition technology is providing unparalleled data and insights into manual assembly lines.
For a large manufacturer, everything is time-sensitive — production deadlines, project deadlines, ship dates, resolution dates, forecasts and employee schedules. The pressure to perform and hit tight timelines is high and constant. Another reality is that no matter how smoothly your production runs, you will encounter problems. Time pressure and production problems can combine to create the perfect storm. Even with the correct systems in place – PDCA and A3 report at the ready – workers and line leadership often struggle to solve problems properly in this context — if they’re solving the right problems at all.
The situation
Imagine you’re a line worker assembling hundreds of medical probes over the course of your shift. You know that your production goal is to move 300 units through your station by the end of the day. You’re on a tight schedule, which is nothing new, but this week your line is running behind schedule because one of your co-workers had a family emergency. You know the team can make up the difference but they’re going to have to work extra hard.
You notice that every once in a while one of the special caps used on your product doesn’t screw on easily. You observe that there are metal pieces in the threads suggesting that there was a machining problem. This is unusual and has never happened before, so in the name of keeping up and not causing a stir, you take a spare adapter and set of pliers and force the cap on, clearing the threads so they can go onto the product. It’s not a big deal, but it is going to add time to each cycle – ensuring the cap can screw on smoothly – making the deadline even harder to achieve. You could notify your line leader to get in touch with engineering to assess the issue, but you don’t want to be blamed if you bring the line to a halt or miss the production deadline.
It’ll be tight, but since this issue hasn’t happened before you decide that you’ll just keep on going with your temporary fix — after all, this will probably never happen again.
Luckily it works out this time. You make the production goal and keep your bosses and customers happy. No reason to bring up the issue as you don’t see any more caps like this toward the end of this production lot.
Three months later…
Everything went smoothly — thankfully no products were returned and the question of a recall was avoided, especially given the risky nature of med device manufacturing. But the next shipment of specialized caps comes in and the same issue occurs from months before. Once again it puts production goals in jeopardy and once again the technician elects to use their workaround.
The point
These issues happen all the time, even with the best intentions from your workers. In the ideal situation, the andon cord should be pulled, the issue reported and solved immediately production will slow until the problem with the cap was resolved and risks mitigated. However, in the real world, we know that there is a gray area where production goals are prioritized and issues are not flagged.
At the very least, the problem should be recorded. In the case of the cap – assuming the cap was an externally sourced part – a complaint and perhaps a supplier corrective action (SCAR) should have been filed with the provider.
What we can all agree is that non-trivial issues like that cited in the example are not for the line associate to assess and solve for themselves.
The solution
To mitigate these kinds of in-line problems that occur during the middle of busy production cycles we need to set up a system by which problems are flagged, documented and available for viewing by those who have the knowledge to fix the issue. In lean manufacturing, PDCA and A3 reports often serve as a way to capture and solve these issues. Even if a work-around solution is needed in the moment, it is a best practice for quality and/or process engineering to document what the change was and whether or not there was any risk to the final product.
The need for these systems seems obvious and is in place in most major manufacturers but it at odds with the realities of time pressures and trust from line workers. Drishti recently conducted a survey surrounding the perceptions of manufacturing from 500 non-management workers. Of the survey respondents with manufacturing experience, only 28% believed that factory workers are treated with respect and trusted by management. This is a telling figure.
So how do we get line workers to participate in the problem solving process, the correct way, every time, as issues occur? We believe it starts with empowering the workers and building trust.
Drishti provides exactly the kind of vigilance, real-time insight and collaboration that can combat these situations as they occur. Using our AI-powered platform our customers (some of the biggest and most lean manufacturers in the world) can see when cycle times are over the expected amount and whether or not standardized work instructions are followed in the right order and/or correctly performed. Think of Drishti in part as an electronic, automatic andon that consistently spots and reports deviations from the standard.
The recorded video is tagged and searchable so that the information about a particular event is never lost in the shuffle again. It is the ultimate warning system with the capability of sending messages to whoever needs the information. More problems will be solved correctly, risks will be more apparent and your workers will feel empowered knowing that they won’t be blamed for issues that are not their fault. Ultimately, production and quality will increase, adding to organizational health and peace of mind for all involved.
Interested in learning more about Drishti’s solution? Learn how Drishti's patented action recognition technology is providing unparalleled data and insights into manual assembly lines.
