Cisco Systems, a company that provides internetworking systems to a worldwide customer base, was experiencing rapid growth. Managers soon realized that top-notch customer service would be a critical component in maintaining Cisco's leadership in the marketplace.

"We needed a system that would allow us to provide the highest level of service support to our partners, distributors, resellers and end users in this rapidly growing environment," says Bill Weidert, service operations manager. Weidert knew that customers' requirements are demanding. "They simply can't afford to be down," he says. "We have a high volume of two- and four-hour response contracts to provide service. We need to have parts available throughout the world that can be accessed immediately and delivered to the customer site."

Cisco turned to LPA Software Inc. (Fairport, N.Y.) to provide the company with inventory planning, demand forecasting, and product and part allocation capabilities. The LPA Supply Chain Solutions (SCS) system orchestrates six major hubs and 200 depots around the world, analyzing inventory, forecasting demand, generating new build orders to replenish supply, and allocating materials to give Cisco's customers the service they require.

Cisco's hardware and software are the foundation of networks that simplify data communications despite boundaries of geography, company structure or in-compatible computer systems. Its products include routers, switches, remote access devices and network management software.

Cisco implemented the LPA software in four months. "Planning was the key to the fast ramp-up," says Earl Stallard, principal of Stallard Enterprises Inc., a software consultation firm that specializes in the LPA system. "We had to create feeds from the old host systems, test interfaces to and from SCS, then take care of all the peripheral processes. Everything that was going to impact the implementation of the software was in a plan, and we followed that plan strictly."

The interfaces to SCS bring in information automatically from around the world on a daily basis.

"Our planners now spend zero time collecting data," Weidert says. "The system automatically collects everything into a single, integrated workspace on the screen — on-hand balances and back orders, stocking levels for remote depots, demand usage history, repair stock available, repair work-in-process, disbursement, actual returns from the field, and vendor order positions for new buy or repair."

Cisco runs the system on the Hewlett-Packard Series 9000 platform using the Oracle database. Eight planners are online at Pentium clients, linked internationally to the global depots across T1 lines and integrated with Oracle Applications for ERP.

Customer orders come in to four Technical Assistance Centers (TAC) manned by Cisco personnel in San Jose, Raleigh, Brussels and Sydney.

"Support is 24-7," says Weidert. "If the customer engineer determines that a part needs to be dispatched, they cut an order into our database. The part is then taken from one of our depots to fix the problem at the customer site."

SCS recognizes that a part was taken and automatically creates an allocation order to replenish it. The part then gets shipped and the inventory stocking level is maintained.

"The system is also forecasting down the line so that we know, as all this activity is occurring, what the aggregate levels should be for any particular part," says Ferdinand Gonzalez, senior service planner. "The system determines requirements at each location at a detailed level so we can forecast ahead and incorporate the necessary lead times. It also knows the manufacturing build time, the repair lead time and the transit times between our main warehouses, hubs and the remote depot that needs the part." That keeps Cisco depots stocked to meet demand.

"The software 'comprehends' our needs on a global basis and at each individual location within the network," Gonzalez says. "It takes all this information into consideration, along with the lead times, and creates orders so a part can be built or repaired and then shipped and received at the depots just in time to meet our customers' requirements."

SCS allocates material replenishment orders on a nightly basis to drive the distribution process. Pick lists are printed around the world in all major Cisco hubs and then distributed to the points of resupply — the local depots in the various countries. The ordering process for a large number of parts has been fully automated, freeing the planners to perform more higher-value-added tasks like analysis.

"The system is geared to do planning by exception," says Stallard. "That means the planners can spend more time on problem parts, rather than just pointing and clicking on parts that have a fairly constant demand."

Cisco plans to automate its planning process even further, especially around the ordering function. "I think we're going to continue to drive toward more automated orders," says Stallard. "The software is set up to run totally automatically, alerting the planners only if they need to look at something."

Brantford, Ontario is known to many as the home of National Hockey League all-time scoring leader Wayne Gretzky. But the Canadian city is also home to Gates Canada Inc., a manufacturer and distributor of hoses, belts and other rubber products for automotive and industrial markets.

In the past few years, Gates Canada has redefined how it approaches demand management and forecasting — and how the company uses these tools to achieve two goals: reducing inventories and improving customer service.

"In the early '90s the service level became key for everybody in our industry," says Vic Portelli, demand manager for Gates Canada. "At that time we had very high inventories and less-than-acceptable customer service levels. Even though service levels were in the low 90s (percentage wise), our customers were demanding improvements. Our back orders were higher than they had ever been, even with all that inventory. So something definitely was amiss."

The problem, of course, was that the company's two goals seemed to contradict each other. Logic says if inventories are reduced, customer service levels may decline as well. But Portelli, who has extensive experience in demand management, production planning and industrial engineering, felt he could create a win-win situation. The key would be improved forecasting. For the answers he needed, he looked to Demand Management Inc. of St. Louis and its Demand Solutions (DS) software.

Less than a year later, Gates Canada improved the accuracy of forecasts and achieved reductions in inventory while improving the quality of service.

"I would say that already our service levels have jumped several percentage points," Portelli says. "Furthermore, we've dropped probably 30 percent on our inventories. I know I can attribute about 15 percent directly to Demand Solutions — half of the reduction. And it's still coming down. As our service levels go up, the number of back orders is coming down, and we have fewer obsolete products."

An affiliate of the Denver-based Gates Rubber Co., Gates Canada has several thousands of products that are sold to nine specific types of industries.

Personalizing the Software
DS enables users to personalize its software, so Portelli had many options for developing forecasts to suit specific types of business.

He could group products by the industry into which they are sold and place products in family groups and subsets of those groups. He could also account for demand levels that vary by geographic region or because of seasonality.

"It is now very easy to look at these different markets," Portelli says. "Since we also break our production down into product groupings and subgroups by family and product type, I thought it was ideal."

Portelli also uses FEEDBACK, a companion program from Demand Management, to create and communicate forecasts that are tailored to the products and demand levels of the geographic regions Gates Canada serves.

"The aggregation features allow me to do a lot of things that were impossible before," Portelli says. "We use FEEDBACK to involve our marketing group in the forecasting process. I aggregate by market, by product type and by subgroup, and send this information to them by e-mail. They have a look at it using their copy of FEEDBACK, enter proposed changes directly on the FEEDBACK screen and return it to me, again via e-mail. FEEDBACK is smart enough to only send back the modified records."

Forecasting Accurately
DS uses 20 different strategies to create forecasts using the most recent periods of known demand history. The resulting statistical forecasts are all easy to understand and, in many cases, mirror what business people already do intuitively when creating forecasts manually. Because demand history is not the only factor that can influence a forecast, DS also allows users to adjust forecasts for outside influences and future events that users know will change demand levels for specific items over the next 12 to 36 months.

To further enhance the accuracy of forecasts, DS incorporates four-year curves into its forecasting strategies. Portelli found the curve function to be useful in predicting shifts in demand because of seasonality and when forecasting new products. Portelli also employs the filters function to isolate products that have non-recurring spikes in demand.

"Take the situation where we have a slow-moving product which we are not forecasting, and have little or no demand history," Portelli says. "Then two, three, or five or 100 unanticipated sales of this product occur. The system may then suggest you need 100 pieces each month for the next 12 months. So what I do is run a filter to give me everything that has one month of history and is forecasting quantities over a certain percentage. The filters isolate them for me. I then go through each item so identified, and determine if our marketing group is planning to start selling this product. If they're not, that's fine. I then adjust the forecast down to zero again."

According to Portelli, the ability to adjust forecasts on the fly for any family group or subset, or even a specific product, has been a major plus for the company. "If we get into a back order situation, it may be that the forecast could be off, or that we don't have the proper lead time in manufacturing or in the purchasing end," he says. "So we have to start looking for the cause. Did we have an exceptional demand? Did we take on a new customer? Is lead time correct?

"One of my responsibilities is to bring down inventories. So if inventories start to climb, I'll break down the file by product families first, then by product types," Portelli explains. "Then I'll meet with the purchasing, manufacturing and/or the materials managers to review the evidence. For minor problems, I'll call up the appropriate material schedulers and discuss a specific problem with a particular product.

"If there are back orders, and I notice that for some reason they're not being filled, I ask, 'Why?' Often when I've looked through DS and the host systems, I've found errors in replenishment planning rules; the order point or whatever is triggering at the wrong time," says Portelli. "We do a lot of re-working of products, from one product (if it's a private brand), to another. Many times information, such as cost or lead time, in the host system might not be correct. I also look at inventory value as well as units, and I find errors in (product) costing." Once problems are identified, Portelli says, the software allows him to create an adjusted forecast quickly and easily so that inventories can be maintained at their proper levels.

Implementation
DS runs on DOS, Windows 95 or NT 32-bit PCs, is compatible with most mainframe systems, and can be installed in about 10 minutes and be functional in a matter of days.

"To get a package up and running, we had very few problems," Portelli says. "I found that the importing and the exporting of their files is so easy. I frequently select and export specific products into Excel for my own comparisons."

The Future
With the DS Forecast Management software in place, Portelli says that the company plans additional measures to improve service levels and manage inventory even better.

Gates Canada is considering DS's DSRP Software for vendor-managed inventory projects with major customers. Already Gates has begun using bar coding and plans to convert to a paperless warehouse system in the near future.

Lucas Aerospace Cargo Systems in Brea, Calif., recently finished implementing the Cincom Acquire configuration and sales-automation tool in order to boost productivity in the plant's maintenance repair and operating supplies (MRO) department. In Brea, Lucas Aerospace manufactures and repairs motors and actuators that airlines and air-freight companies use to move cargo on their aircraft.

"We will now see the workload diminish to a point where people can have a life of their own," says Tom Glahe, a 19-year Lucas Aerospace employee, who led the Cincom Acquire implementation at Brea. "A lot of them are working weekends and long hours at night just to maintain the legacy system. We hope to eliminate that."

At this point, Glahe says, Cincom Acquire has produced its biggest time savings in the component configuration and routing of repair jobs. "One employee told me that what she used to do in two and a half hours now takes her 40 minutes," Glahe says. "By taking less time to accomplish these tasks, we can now generate the customer quotes a lot faster, get their approval sooner and get the product out a lot faster — that we have already seen."

According to Glahe, the Brea IS staff decided to implement Cincom Acquire when they determined that they could implement the Cincom product cheaper and faster than they could modify either their legacy MRO system or their full CONTROL system.

Glahe also plans to use the new Cincom system to place the plant's component maintenance manuals online, in order to speed some of the certifications required by the U.S. Federal Aviation Administration. "We hope to have Cincom Acquire generate the costing and help in the engineering logistics," Glahe says. "We hope to be able to give that tool to engineering so they'll be able to improve our product."

Artex International, a Highland, Ill.-based manufacturer of tablecloths and napkins, suffered from high employee turnover in its St. George, Utah, plant. Productivity declined and training costs were burdensome. An extremely low unemployment rate in the area — less than 3 percent — exacerbated the problem.

Plant manager Jim Duby responded with an innovative plan. His goal was to discern the major behavior and personality traits of the plant's successful, long-term employees, and then attempt to hire new workers with the same characteristics. "We realized that associates who stayed on board for six months were staying for the long term," Duby says. "But we were churning new employees, some who stayed less than a month."

The results of the plan were dramatic. In less than one year, Artex reduced turnover by half in the 60-worker St. George plant and cut turnover-related costs by $31,000. Key to Duby's strategy was an organizational management tool called the Predictive Index (PI), a survey that yields a profile of a worker's personality traits and aptitude for certain types of work. It was through the PI that Duby and his supervisors learned that their plant associates shared a similar behavior trait — a High D Factor. "High D" individuals like clear information and strong, inspiring leadership from their managers.

Developed in 1955 by Arnold S. Daniels, founder of Praendex Inc. of Wellesley, Mass., and based on the work of noted behaviorist B.F. Skinner and others, the PI suggests how managers can more effectively communicate with workers, and has the potential to dramatically improve productivity and profitability. Through Praendex's PI Management Resources, an association of consultants who train clients in the use and interpretation of the PI, companies can implement the PI throughout their organizational structures.

Duby and his plant supervisors first took the PI themselves and became trained in interpreting the surveys. Then Duby took a critical step in overcoming employee discomfort with the new management tool. "We posted the supervisors' results out in the plant," he says. "That took away any fear there might have been."

The PI takes less than 10 minutes for most users to fill out. It consists of a single sheet of paper with identical lists of 86 words on each side. On the first side, users are instructed to check a box beside each word that describes how they believe others expect them to act. On the second side, users are instructed to check a box beside each word that they believe really describes them.

Managers or human resources professionals within the company who have been trained in how to evaluate the PI then review the results and determine what type of position a potential new employee may thrive in, or whether the prospective hire would be appropriate for the job that is open. "If you understand the behavior traits of people you work with, you can create an environment where they thrive, and the company does too," Duby says. "With the PI surveys, we started seeing patterns, especially among the people who had been with us for a while. We saw people with a high-level of patience, and they also showed a high-degree of wanting to be in an atmosphere with set procedures and rules. It became clear why we failed with the people who had left, and we realized we should be looking for these characteristics in new hires."

The results of the PI, both from long-term associates and new hires, enabled Artex to implement plans for improvement in several management areas, including turnover, communication and motivation, supervision and training, conflict resolution and morale building. Duby implemented new human resources procedures that require all new job applicants to take the PI survey as a pre-screening measure. All applicants' PI surveys are kept on file, but only those with the key traits that suggest a potential for success in a certain job are invited to interview for openings. Additionally, when a job opening occurs, Duby and his staff examine the PI surveys they already have on hand to see if a previous applicant should be offered an interview.

Knowing the importance of clear communication with their High D associates, Duby and other plant managers and supervisors made it a point to meet with each associate to communicate matters important to the company and to individuals. This kind of communication has resulted in a much more informed and satisfied work force.

"The most common problem with communications is the absence of them," Duby says. This causes uncertainty, which causes low morale and negative attitude. Morale and attitude have a direct effect on productivity and profits. Our people now put more energy into their jobs, have a greater sense of involvement and responsibility and get more satisfaction from their work."

Though he initially utilized the PI to reduce turnover among recent hires, Duby notes that the survey and the advice from their PI Management Resources' consultant had unexpected benefits within Artex's St. George plant.

"We found that the PI is also a very useful tool in conflict resolution. It is important to focus on the behavior that creates a problem, not the person," says Duby. "I have frequently found that when the problem behavior occurs, it is due primarily to the unfulfilled needs of an associate rather than being due to a 'bad apple' issue.

Copyright © 2020 by APICS — The Educational Society for Resource Management. All rights reserved.

Web Site © Copyright 2020 by Lionheart Publishing, Inc.
All rights reserved.


Lionheart Publishing, Inc.
2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA
Phone: +44 23 8110 3411 |
E-mail:
Web: www.lionheartpub.com


Web Design by Premier Web Designs
E-mail: [email protected]