New Tool: ProSource
Checkout our packaging and processing solutions finder, ProSource.

Calculating KPIs in real time

A leading baked goods and snacks manufacturer located in the United Kingdom has been a longtime customer of ICONICS, a member of the Control System Integrators Association (CSIA). 

Years ago, ICONICS implemented automated controls for multiple factors (oven control, product quality monitoring, etc.) within a few of the manufacturer’s product lines to help ensure high quality while simultaneously improving line efficiency. That initial project involved installation of multiple ICONICS software solutions for data visualization and control, as well as for OPC data aggregation, bridging, redundancy and tunneling.

Recently, the same baked goods and snack manufacturer merged with a global biscuit and confectionery company. Now a subsidiary company of this newly formed enterprise, the processor wanted to further its monitoring and control abilities. By this time, the number of manufacturing product lines had increased, which can be handled by the latest generation of ICONICS software, including solutions such as the GENESIS64 HMI and supervisory control and data acquisition (SCADA) suite and Hyper Historian high-speed, robust data historian, according to Melissa Topp, senior director of global marketing for ICONICS.

Some of the new key performance indicators (KPI) provided by the current ICONICS solutions include:

• Shift output (in kilograms).

• Line production speed (in kilograms per hour).

• Shift waste (in kilograms).

• Process line downtime.

• Downtime (shift).

• Line turn of scale (ToS) (in kilograms).

• ToS (in percentage).

• Shaper/cutter speed (in revolutions per minute).

• Previous shift output (in kilograms).

• Previous shift waste.

• Previous shift downtime.

• Previous shift ToS.

•Case count (last 24 hours).

Of particular importance were the addition of turn-of-scale KPIs, helping to determine any giveaway weight. Turn of scale refers to the giveaway weight of a product (e.g., a single package of a specific baked good or snack).

According to Topp, there are two distinct inputs, which are monitored and measured in real time: 

• Total weight produced in the specific time frame. 

• Total packages accepted in the specific time frame.

“The number of outright rejected packages is not included in the calculation of the total weight,” states Topp. “The manufacturer is leveraging ICONICS Hyper Historian to calculate and record the division of total weights/packages accepted, and then records that information as a snapshot of weights for 30 minutes.”

There is another raw input called referenced weight, which is the target weight for the product. This value is stored in the ERP/SAP system. “Due to the fact that ICONICS uses standard industry connectors, it was very simple to extract this data,” states Topp. “The solution performs a subtraction between the reference weight and actual weight and then expresses it as a percentage difference in comparison to the reference value.”

Topp provides an example with real numbers. Imagine that between 5:00 p.m. and 5:30 p.m., the total weight recorded was 2,050 g and 10 packages were accepted. The average package weight will be 205 g (2,050/10). The reference weight happens to be 200 g for that specific SKU. The percentage difference is 2.5 percent turn of scale or “giveaway” [((205-200)/200) *100]. ICONICS Hyper Historian provides an average of percentages during a product manufacturing run so that the manufacturer can see the value trending.

The baked goods and snacks manufacturer can generate a live report containing a trend view of runtime status over the past 24 hours. The same report can also contain current shift metrics (including shift output [in kilograms], downtime [in minutes], shift waste [in kilograms] and shift output [in kilograms]) in comparison to previous shift (for the same data points). “The ability to have these KPIs calculated in real time allows operators and supervisiors to make immediate decisions and to be much more proactive when monitoring data (i.e., saving money on associated production costs wherever possible),” Topp says.