Present time and motion studies

In today’s manufacturing environment time and motion studies are conducted manually using stopwatches. When dealing with processing times that are in the range of seconds to minutes, an observer (often a manufacturing or industrial engineer) simply “peeks over the shoulder” of the employee to create a sample of observations. This data is then stored in a database (usually Microsoft Excel), manually analyzed by an engineer over multiple days and used as a benchmark for lean improvement efforts. The main advantage of using this approach is its simplicity and low cost however, as we’ll see below there are plenty of problems associated with this approach.

Data measured using the current approach is biased and inaccurate

The biggest challenge with the current manual approach for measuring labor productivity is that the data produced is highly biased and doesn’t represent the true picture of the production environment. That is because these manual time studies are usually conducted once or twice per year for a couple of hours on a given day. Manufacturers usually produce for more than 50 weeks per year so a sample size of a few hours on a couple of days does not represent the entire production environment.

As a result of this practice, the data only represents a short period of production and oftentimes these are the less busy days because on busier days it is tough to get these studies done due to production priorities. Furthermore, workers tend to work extra carefully when being monitored by another person and this behavioral change impacts the data even further. In essence, the current method of measuring labor productivity using stop watches produces data that is unusable as it provides minimal insights into the entire production operation.

Computer vision can conduct time and motion studies continuously

Computer vision provides non-obstructive data collection in real-time for all workers, in addition to spotting specific opportunities for improving the processes. Cameras installed on the manufacturing floor capture activities such as the assembly of a car engine. Deep learning methods help read the video data and identify objects, like an engine assembly or an operator in a production plant. The computer vision software then uses a set of rules to determine processing times. This computer vision software enables the generation of real-time reports to management for taking immediate action.

This computer vision powered approach is far superior to manual time studies because the system is constantly running and so is able to get all the production data with high accuracy. Furthermore, the results are produced instantaneously without any manual data wrangling which allows the manufacturing operations team to primarily focus on improvement efforts and gaining value quickly.

How i-5O clients are increasing revenue using computer vision

i-5O has been working with manufacturing clients helping them conduct continuous time and motion studies using our vision system. One of i-5O’s long-term clients, Pride Mobility uses our vision system to track the products that are in waiting for a period of time that exceeded set thresholds determined by management. A notification system was implemented in which if the products exceed this threshold, an alert is sent to management personnel. This allows management to address any issues with the chair immediately without causing significant delays in the production timeline and keeping the production flow consistent. As a direct result of this, Pride Mobility has been able to generate an additional 26% in revenue on a monthly basis.

Using computer vision to increase

labor productivity in manufacturing

BY KHIZER HAYAT

5 MINUTE READ

Present time and motion studies

Data measured using the current approach is biased and inaccurate

Computer vision can conduct time and motion studies continuously

How i-5O clients are increasing revenue using computer vision

Try it for Yourself