Best Manufacturing Practices
This year, Food Engineering surveyed subscribers selected randomly from our circulation list. Of the 132 respondents to our survey, 59 percent have more than 20 years of experience in the food industry, 30 percent have 10 to 19 years of experience. Ninety-two percent work at manufacturing locations, 8 percent at non-manufacturing (corporate headquarters or R&D) locations. (See box for more information about respondents.)
We asked respondents to rate the improvements in manufacturing efficiency which their individual plants and companies have achieved over the past five years according to seven basic measures:
- Percentage of reduction in manufacturing costs; manufacturing cycle times; inventories (both raw materials and finished products); order lead times; and plant waste.
- Percentage of increase in man-hour productivity and throughput.
We found that 80 percent of our respondents measure manufacturing efficiency in terms of throughput (volume or cases-per-manhour); 70 percent in terms of waste reduction; 55 percent by improved safety/lost-time records; 54 percent by reduced customer complaints.
We were rather surprised to find, however, relatively low percentages who measure other key performance indicators (KPIs) such as:
- Conformity to production schedules (53 percent);
- Changeover time (37 percent);
- Cycle-time reduction (21 percent);
- Order lead-time reduction (20 percent).
Last year, our "knee-jerk" reaction was that these KPIs were probably applied only by larger plants, but close examination revealed that those plants which apply these measures range across all plant sizes. The same seems true this year. "Other" performance measures cited by respondents include yield, downtime, and quality measurements.
As shown on the adjoining table, respondents report reducing manufacturing costs at their individual plants by a mean of 12.4 percent and company-wide by a mean of 15.2 percent over the past five years. (A mean is the midpoint; half of the reported percentages are above, half below.) These figures are substantially improved over the mean 7.1 and 6.4 percentages reported last year, indicating remarkable recent year-to-year improvements.
Most respondents (53 percent) report plant manufacturing-cost reductions of 1 to 10 percent over the period; 28 percent report reducing their plant costs between 11 and 20 percent. Only 3 percent report no reduction in manufacturing costs, while 7.4 percent reduced plant costs by 21 to 30 and 31 to 40 percent respectively.
Those reporting manufacturing cycle-time results show mean improvements of 17.2 percent at plant level and 16.2 percent company-wide -- well up from the mean of 9.7 percent reported in both categories last year. Nearly one-third (32.7 percent) reduced cycle times by 11-20 percent; another third reduced cycle times between 1 and 10 percent.
The most dramatic improvements, however, show-up in reduced inventories of both raw materials and finished products. Respondents report reducing raw-material inventories by a mean of 16.7 percent at their plants and 18.8 percent companywide as compared to 6.7 and 5.9 percent last year. As shown, most (slightly more than 50 percent) reduced raw-material inventories at both levels between 1 and 10 percent, but nearly one-fifth (19 percent) report reducing raw-material inventories companywide by 40 to 50 percent!
Finished-goods inventory improvements are somewhat lower, but still well improved over last year by mean percentages of 12.3 percent vs 8.6 percent at plant level, and especially of 14.3 vs 6.7 percent companywide.
Of the relatively few respondents (20 percent) measuring order lead-time reduction, the mean reduction at plant level tripled to 22 percent as compared to 7 percent last year, and improved companywide from 8 to 13 percent.
As noted above, respondents report improved man-hour productivity and throughput but at lower levels than reported last year, suggesting that most improvements occurred earlier during the five-year period surveyed. Mean man-hour productivity at plant level, for example, was reported up 13.7 percent this year vs 19.6 percent last year; mean improvement in plant throughput at 14.8 percent vs 22 percent in 2000.
Training, equipment and control
We also asked respondents to list the factors contributing to their improved manufacturing results. As reported last year, improved training, equipment and control ranked as the top three factors, with "improved employee/operator training" cited by 79 percent of our respondents just ahead of "new equipment " at 78 percent. Improved automation and process control ranked third at 72 percent.
Although the same percentage of respondents (42 percent) reported "improved maintenance systems" this year as last, maintenance jumped from seventh to fourth place among the reasons for improved performance. Improved workforce structures, although cited by 41 percent of our respondents both this year and last, moved-up from eighth to fifth place among factors listed. Those citing a "new or renovated plant" fell from 43 to 17 percent, suggesting a decline last year in new projects (to be confirmed or contradicted in April by FE's annual "Food Plant Construction Report.")
Only 31 percent of respondents cited "improved manufacturing flexibility" as compared to 53 percent last year, but this reflects the latest five-year period and does not reflect a decline in flexibility.
Although improved automation and process control ranked third among the reasons for improved manufacturing performance, plant-floor integration -- as shown on the pie chart -- shows little change from last year, when substantial improvements were reported. Forty-four percent of respondents, for example, report "scattered islands of control" versus 42 percent last year; 8.6 percent report "top to bottom integration" as compared to 8 percent last year.
When asked what types of process controls are installed in their plants, respondents replied as follows: PLCs, 68 percent; batch controls, 53 percent; individual microprocessor-based machine controls, 53 percent; maintenance software, 40 percent; material/manufacturing requirements (MRP) software, 28 percent; statistical process control (SPC) software,14 percent; quality-analysis/laboratory information-management (QA/LIMS) software, 14 percent; ERP software, 9 percent (surprisingly low); SCADA, 6 percent; predictive control, 5 percent, manufacturing execution systems (MES), 4 percent. The relatively high percentage of individual OEM machine controls seems to account for the relatively high "scattered islands of control."
When respondents were asked how far up the integration ladder their companies have networked to date, 46 report integration at the device level as compared to 44 percent last year. But respondents seemed confused -- or perhaps we confused them -- when asked about integration with information systems. In response to the same question, 19 percent replied integrating via Internet with customers and suppliers. When later asked how far their companies have gone in managing the supply chain, only 12 percent reported e-commerce integration with customers and suppliers via Internet.
This apparent contradiction may reflect corporate priorities to integrate IT systems with each other rather than with plant systems. According to the Bureau of Economic Analysis, as reported last month by FE (Jan., '01), capital investment for IT equipment has surpassed spending for non-IT equipment. Currently, only 5 percent of IT investment is spent on manufacturing execution and asset management.
Incremental improvement in process control is leading to more than incremental improvement in manufacturing flexibility. This year, 66 percent of respondents report "great flexibility" in product type, 54 percent "great flexibility" in product size, and 55 percent "great flexibility" in packaging as compared to last year's results of 54, 48 and 44 percent respectively. Only 11 percent report "little or no flexibility" in packaging versus 14 percent last year.
Maintenance: puzzling picture
Some maintenance results fit last year's trends, others seem way off. We'll accept this year's picture as more accurate since there are more respondents.
Last year, 71 percent of respondents reported preventive maintenance systems, close to the 67 percent reporting preventive maintenance this year.
This year, however, only 43 percent report maintenance as a separate department versus 64 percent last year. If maintenance is truly being decentralized, it doesn't square with other results. Last year, 33.9 percent of respondents reported that operators troubleshoot and perform routine maintenance tasks, essentially the same as 33.3 percent this year. Last year, 27 percent reported maintenance incorporated into production-line teams versus only 11 percent this year.
Another puzzle: Last year, 39 percent of respondents reported using computerized maintenance management systems; this year only 21 percent have CMMS. More respondents to this year's survey may account for the difference.
Thirty-nine percent of respondents are stuck in a reactive-maintenance mode, but 23 percent have installed predictive-maintenance systems, up slightly from 20 percent last year.
More outsourced services
Respondents were asked to estimate the percentage of several services outsourced by their companies. As compared to last year, they cite more outsourcing of engineering, maintenance, microbiological testing and energy management.
Respondents report outsourcing a mean 45 percent of engineering services, substantially up from 31 percent last year. The largest group of respondents, 22.6 percent, outsource 1 to 10 percent of their engineering work, but 17 percent of respondents outsource up to 50 percent of engineering and another 17 percent outsource it all -- 100 percent!
This year, respondents report outsourcing a mean 17 percent of maintenance work as compared to 9 percent last year. Microbiological testing is outsourced at a mean 42 percent versus 27 percent last year, reflecting food-safety concerns. Energy management is outsourced at a mean 18.6 percent versus 8 percent last year, reflecting growing use of energy service providers (ESPs).
Food-safety: training & HACCP
When asked what methods their plants have implemented to better assure plant sanitation and food safety, respondents closely mirror last year's findings. Again, improved training (77 percent), HACCP (73 percent), and regular review of GMPs, SOPs and SSOPs (63 percent) top the list. Two slight improvements are seen: This year, 12 percent report a HACCP coordinator with no other duties versus 9 percent last year; only 2 percent report applying "none" of the 10 food-safety initiatives listed, down from 6 percent last year.
Little change in workforce design
Nearly half of our respondents (46 percent) report no change in the traditional "command-and-control" management structure of their workforces. Twenty-two percent have established management-directed or shared-leadership teams on the plant floor; 19 percent have self-directed work teams.
Of those respondents whose plants have implemented self-directed work teams, 76 percent say the concept works. In companies with shared-leadership or self-directed work teams, 71 percent of the teams make on-line process-control decisions; 64 percent make on-line quality-control decisions; 40 percent participate in interviewing and hiring new team members; and 29 percent make purchasing decisions. This year as last year, 5.5 percent of respondent companies partner with unions in plant management.
In addition to workforce-design initiatives listed on the adjoining table, one respondent cited "a heavy reliance on temporary labor to supplement a very unstable labor pool."
The methods used to train and qualify employees vary little from those reported last year. Not surprisingly, supervised on-the-job training is most common, as reported by 93 percent of our respondents. Fifty-five percent cross-train employees in various skills, but -- although up 5 percent from last year -- only 23 percent have pay-for-skills programs, and only 11 percent have established skill-block training and certification programs. Forty-three percent utilize vendor-training programs; 43 percent conduct on-site classroom training, and 27 percent utilize off-site training opportunities at community colleges and technical institutes.
This year, in light of the tight labor market, we asked what methods companies use to recruit and retain good employees.Respondents listed 18 different incentives.
Sixty percent of our respondents report that their companies offer 401(k) savings/investment programs; 56 percent have improved medical/dental benefits; 55 percent offer bonus opportunities.
Most respondents listed improved training opportunities as incentives to recruiting and retention: 36 percent cite tuition-reimbursement programs, 36 percent report pay-for-skills programs (contradicting the 23 percent reported above); 23 percent cite improved opportunities for technical/professional training; 19 percent cite fundamental skills training; another 19 percent cite team-training. Another 19 percent, too low, report that their companies demonstrate "trust, real empowerment" as an incentive to retaining people.
As shown on the table, "lack of qualified people" and "justifying the cost" are seen by our respondents as the greatest barriers to implementing "best practices" in their plants. In addition to the barriers listed, one respondent cited "union contract."
Wide range of 'best practices' To round-out our survey, we asked: "What two or three 'best practices' have been most effective in improving manufacturing efficiency at your plant?"
This generated 80 responses covering a wide range of practices. The largest group of 24 responses (30 percent) addresses "people issues" such as improved training, incentives and communications. Among them: AIB training; a bonus program; empowerment; ownership; problem-solving and process-improvement teams and "total communication between floor-level supervisors and upper management."
The second largest group, totaling 16 responses (20 percent), can be classified as improved management practices. Examples: "benchmarking to determine better ways of achieving the same goals;" a cost-improvement program to reduce SKUs; "focusing on customer complaints;" a focus on "the highest-cost items" offering the "easiest pickings;" problem-solving and process-improvement teams.
Ten responses (12.5 percent) relate to improved automation or information systems, including automated collection of process data; "capacity planning;" line-tracking; ERP for a "common information system."
Nine answers cite improved engineering and equipment, such as high-tech equipment "simple to operate and maintain;" multiple plants using the same equipment and suppliers to reduce parts and share knowledge. Eight responses relate to improved maintenance, including "improved scheduling of maintenance shifts;" preventive maintenance CMMS; "expanded predictive maintenance." Another eight responses address improved food safety and quality assurance. Examples: "HACCP plan installed;" "QA at the individual employee level."
To conclude our survey, we asked: "What two or three 'best practices' not currently implemented would you like to see implemented in your plant?"
This sparked 51 responses, nearly half (24) relating to workforce issues. Typical answers: "Better qualified people;" "better training of operators;" matching experienced employees with others for better cross-training; "more classroom education;" "profit sharing;" "self-directed work teams;" TQM for "employee involvement and empowerment."
Eleven respondents (22 percent) hope for improved automation or information systems, such as "greater computer integration;" "improved supply chain;" "more info and accountability for operators;" "better measurement of work with less paperwork" and "expanded knowledge and use of ERP." The third-largest response group covers engineering and equipment desires. Examples: "better process flow and facility design;" cost-justification and design of "clean room;" more continuous processing "to reduce energy costs."
One respondent has realized his hope: "A pay-for-skills practice has just been implemented!" he reports.