| |
|
The need to integrate more readily with customers and pressures to control costs are among the top reasons driving mills to implement new information systems, according to a 1999 Pulp & Paper survey. In addition to adding new systems, many mills are replacing legacy systems because of Y2K compliance issues. Although the majority of mills responding reported that they are pleased with their new systems from an investment standpoint, most respondents were guarded about the wide-ranging impacts these systems could have on business practices and insight into supply chain management.
To determine the types of information systems in North American mills and how they are used, Pulp & Paper conducted an exclusive survey that asked for information about hardware and software associated with such systems throughout the mill. Survey respondents were either information systems managers at the mill or corporate level, or they were mill managers. Almost every grade, geographical region tonnage range, and mill size were represented in the responses.
Survey responses indicated a wide variation in the level of information technology implementation at mills. Many larger mills appeared to have a variety of tightly integrated systems that share data automatically, and such mills usually worked across a variety of platforms (PC, mini- and midi-computers, and mainframes). They also tended to have a high PC-to-employee ratio. However, many smaller mills had one or no information systems and worked from a mini-computer or mainframe alone. In the 1999 survey, a higher number of these smaller mills responded, which apparently caused the percentage of reported information systems to drop somewhat across a variety of software categories when compared with the 1998 survey.
COMPUTER HARDWARE AND NETWORKS. Pulp & Paper’s information systems survey asked specific questions about the types of networks existing in mills and how they are used. Survey participants were also questioned about the hardware existing at their mills, as well as about plans for upgrading from mainframe computers to PCs.
Networks. Although a higher percentage of smaller mills responded to this year’s survey, the number of IS managers reporting LANs throughout their mills increased from 88% to 92% (Figure 1). However, the incidence of WANs dropped somewhat from 81% to 70%. The lack of a WAN seemed more related to the product rather than yearly tonnage or total employees. Pulp mills were the least likely to have such a network and made up 30% of those without a WAN, followed by small tissue mills and board mills of various sizes.
Mills with a WAN were asked to identify what applications the network was used for other than e-mail. Although reported less frequently than in the 1998 survey (61%), business and financial systems (42%) that handled payroll, accounting, general ledger, and administrative functions were still the most commonly listed. The second most common answer—a result that did not appear in the 1999 survey—was access to various mill software applications and information systems (26%) such as an ERP system or CMMS.
Order fulfillment, shipping, and inventory applications (23%), along with interfacing with sales (14%) and production (9%), were also common uses for a WAN. Other answers included remote system administration, Internet and Intranet access, data transfer and file sharing, and materials safety data sheets (MSDS).
Internet use. Ninety-five percent of the mills responding indicated that they have an Internet connection—a 14% increase since the 1998 survey. When asked what applications involving customers and suppliers the Internet was used for, common answers included email (12%), connecting with suppliers, their web pages, and their technical support (10%), and customer service applications such as complaint follow-up (7%).
Around 34% of respondents either listed no answer or specifically indicated that they did not use the Internet for customer/supplier applications. However, about 20% of these answered “none yet,” indicating some expectation for future use in those areas.
According to G. Guy Sorrell, a partner with Deloitte Consulting/ICS, a provider of business consulting and systems integration services, an often cited but rapidly disappearing deterrent to increased Internet usage has been a lack of qualified users. However, Sorrell predicts that “Internet usage will grow exponentially over the next few years and customer/supplier transactions will become a strongly emphasized application area.”
Hardware. Most mills have a combination of PCs and mini- or midi-computers (57%), while another 11% include a mainframe in the mix. PC networks alone (15%) were less common and appeared mostly in mills with less than 200 employees. However, when asked what computer the mill primarily uses, PCs alone (34%) and PCs and minis or midis (34%) tied for the most prevalent. Another 5% reported using mainframes along with the PCs, midis, and minis. These findings are in line with last year’s responses, where 37% reported a primary use of PCs and 42% a combination of platforms.
Providing centralized and easy software administration, mid-range platforms such as the IBM AS/400 appear to have grown in use according to the survey. However, the power of PCs for the individual user has not been overlooked. As with last year’s survey (90%), the majority of respondents (85%) indicated that they have PCs on the mill floor.
An interesting statistic was revealed in the 1999 survey related to the number of PCs and the number of employees at mills. At most mills, there was a PC for one in every three employees. However, this number was as high three in every four employees at one large mill, while some smaller mills relied on only a mini-computer or mainframe.
Sixty-nine percent of respondents said that their current hardware setup was adequate for the needs of their mills. However, 33% indicated that they were planning upgrades for the future. The most common upgrade listed was a LAN expansion or upgrade, followed by the addition of more PCs, new servers, and the move to an AS/400 environment.
FIGURE 1: Ninety-five percent of the mills indicated that they have an Internet connection—a 14% increase since the 1998 survey.
CLIENT/SERVER TECHNOLOGY. With the proliferation of off-the-shelf client/server packages, mills have become more aware of the functional and cost benefits of these systems. Eighty-five percent of IS managers surveyed were familiar with client/server technology, which represents a nine point drop from last year’s survey. However, almost all mills unfamiliar with this technology in the 1999 survey were smaller, producing less than 150,000 tpy.
Of those mills familiar with client/server systems, 50% indicated that advances in that technology had led them to examine more software for use in the mill. The most common reasons for implementing client/server systems was that they are user-friendly and easy to maintain for system administrators. For example, a respondent from a Midwestern linerboard mill said this technology allows the IS department to “maintain one server software set for the entire mill.”
Other common reasons for adopting client/server technology included customer requirements, low system cost, flexibility of the software, better data reporting, reduced dependency on programmers, Y2K compliance, the availability of off-the-shelf standard software, and access to more and better information. One respondent described the benefits of “having production data at your fingertips.”
About 20% of respondents familiar with client/server technology indicated that advances in those systems had not necessarily led them to consider more software for their mills. This response was in line with last year’s (23%) and comments again indicated that purchasing decisions were usually need-driven rather than technology driven. For example, one respondent claimed that “client/server technology does not drive our decisions or requirements.” Another respondent said the technology had not driven further software examination, but conceded that “it had made a wider variety of situations possible.”
Other respondents were not so enthused with client/server systems for technological reasons. A respondent from a larger Northwestern mill said that they were “leaning toward web-enabled systems with JAVA and ASP.” For more information on this technology, see the sidebar, “Web-enabled systems: The Next Step?”
Relational databases. To provide object-oriented power and functionality, client/server systems depend on relational databases. Sixty-four percent of respondents in this year’s survey reported having such a database, which is down significantly from last year’s figure of 85%, again reflecting the higher number of respondents from mills producing less than 150,000 tpy.
Oracle (44%) was still the most common relational database, followed by Microsoft SQL Server (41%), which jumped 12 points from last year. Informix was again in third place (18%) and was joined by IBM DB2/400 (18%), which did not figure as prominently in last year’s responses. Informix was also mentioned (10%). In addition, some mills indicated that they had developed their own relational databases.
INFORMATION SYSTEMS IN THE MILL. The survey asked questions about the existence of seven different types of information systems within the mill. For each of these seven categories, survey participants were asked how the data from this system was primarily used. These systems were:
• Computerized maintenance management system (CMMS)
• Advanced planning and scheduling system (APS)
• Manufacturing execution system (MES)
• Process control software
• Warehouse management system (WMS)
• Miscellaneous systems
• Enterprise resource planning (ERP) systems
In this year’s survey, the percentage of mills having information systems dropped in each category except process control software. This drop appears to be associated with the higher number of small mills that responded this year, since many of these had only one or no information systems. Figure 2 shows the types of information systems found in mills and how common they are, based on survey results.
Survey responses indicated a wide range of functionality within each category, and some software packages were listed in more than one category. This reflects the lack of functional boundaries in software packages that are currently available (see the sidebar, “Open Systems Blur Software Category Boundaries”).
CMMS. While the CMMS was the most common mill information system listed in the 1998 survey (79%), it dropped to second place in this year’s survey, with 54% of mills reporting such a system. As in last year’s survey, most of the packages mentioned were vendor-supplied (64%) rather than in-house developed. Various CMMS packages supplied by Marcam Solutions Inc. were again the most common. Other packages mentioned included those by Indus, J.D. Edwards, PSDI, and Walker Inc. SAP’s maintenance module—a part of its ERP solution—was also listed in this category.
The CMMS, along with process-control software, again appeared to have the most clearly defined functionality, as indicated by consistent answers about CMMS use. The primary uses of CMMS data noted by survey respondents included work order management (24%), maintenance planning and scheduling (24%), inventory control (15%), and preventive/predictive maintenance activities (15%). Other responses included purchasing, tracking of equipment history, reporting, and decision making.
APS. Thirty-three percent of survey respondents indicated that their mills had an advanced planning and scheduling system (APS), which is up somewhat from last year (27%). As with the CMMS, the majority (60%) of these were vendor-supplied. Vendors included Majiq, Measurex, Microsoft Project, Indus, and IBM.
Responses indicated that the APS is most commonly used for production and machine scheduling (45%), including shutdown scheduling. Other responses included trim/load planning, invoice tracking, and decision making.
To Tom Bruhn, senior vice president and general manager of Tava Consulting, a manufacturing consulting and systems integration firm, the responses in the APS category indicate that there is a lack of understanding within the paper industry regarding supply chain management software. According to Bruhn, none of the top 25 supply chain software vendors are cited, and, of those noted, none are really considered APS vendors. However, Bruhn feels that the paper industry probably has more to gain from APS than most industries today.
“Advanced planning systems can provide scheduling personnel with the information to assess numerous options at the time an order is placed,” explains Bruhn. “For example, with the right tools, the planner can assess the cost difference between lower machine utilization versus less finished goods. The paper industry notoriously optimizes local functions to the detriment of a global optimum. APS provides the ‘big picture,’ and can offer this level of system support simultaneously across multiple plants.”
FIGURE 2: The survey shows that some information system types are more common than others.
MES. Manufacturing execution systems were reported by 28% of survey participants in the 1999 information systems survey, which is down somewhat from the 37% reported last year. Other than process control software, the MES was the most likely system to be vendor-supplied (76%) rather than in-house written. The most common vendor mentioned was Measurex (and its OptiVision product) at 35%. Other vendor/products included Majiq, OSI (PI System), Mountain Systems (Gradebook), Indus (EMPAC), ABB (Autoprod), and Peoplesoft.
Survey participants reported a wide variety of uses for an MES. Inventory control and tracking, production reporting, shipping, billing, quality control, order entry, process monitoring and control, decision making, and sales tracking were among the answers.
Process control software. The most prevalent information system reported in this year’s survey was process control software (84%), which rose from second place in the 1998 survey (69%). This category includes both traditional distributed control systems as well as next-tier information software that ties production-related data together for a more millwide process and production overview. Virtually all of these systems were vendor-supplied, and the reported vendors ranged from Measurex, Honeywell, ABB, Wonderware, Fisher, Bailey, Foxboro, Moore, and Valmet to OSI’s PI system.
The most frequently listed use for process control software was monitoring and control of the papermaking process and operations (35%), which was followed by process trending and tracking (12%). Other uses included troubleshooting, reporting, quality control, and decision making.
WMS. Forty-one percent of survey respondents indicated that their mills use a warehouse management system, which is down somewhat from last year (48%). As with the other information system categories, the majority of these systems were vendor supplied. The most commonly mentioned vendor in this category was Majiq (24%), followed by various Measurex systems (12%), including OptiVision and PMIS. Other vendors/products included Indus, Oracle, Bonn, Valmet Automation, and MOPS.
Inventory control (28%) was the most commonly listed use for a WMS, followed by roll tracking and shipping. Other uses mentioned were quality control, customer service, decision making, production scheduling, and storage control.
FIGURE 3: This year’s respondents were more guarded in their opinions about new information systems.
Other information systems. Respondents were also asked to identify any mill information systems other than those noted in the previous sections, and 36% reported such systems, which was down 8 points from last year. Unlike last year’s responses, which indicated almost an equal number of specialized production systems and business systems, this year’s answers were dominated by business software (55%), including accounting, general ledger, and payroll software. In addition, most of the systems in this category appeared to be vendor supplied, which was not the case last year when almost half of the software was described as in-house developed.
Additional responses included OSI’s PI software used as a millwide information system (18%), followed by quality tracking and support software (14%). Purchasing, stores, inventory, laboratory, preventive maintenance, order management, and manufacturing software were also listed.
ERP, financial, and corporate systems. Only 16% of mills responding to the survey indicated that their mill or corporation had an ERP system, which was down significantly from last year’s 31%. In addition, only 3% reported that they were evaluating the purchase of an ERP as opposed to last year’s 10%. As with the lower percentages reported in the other categories, the decrease in the number of ERP systems reported appears associated with the increase in responses from smaller mills.
Almost all of the ERP systems mentioned were vendor-supplied. Common vendors included Peoplesoft, J.D. Edwards, SAP, Measurex, and BPCS.
Survey respondents were also asked to list the top three reasons for purchasing an ERP. The most common responses were Y2K compliance issues, order management/entry capabilities, level of database and system integration, cost of the system, and financials. Other answers included customization features, tracking of work in progress (WIP) and inventory, flexibility, control of manufacturing with bill of materials (BOM), and the vendor’s market share. However, one respondent described the purchase as a “highly placed decision by poorly informed executives.”
Integration of multiple systems. Pulp & Paper’s survey asked respondents with more than one of the previously described systems if those in any way share data or work from the same database. Thirty-nine percent of respondents said “yes,” while 34% answered a definitive “no.” Compared with last year, when 58% responded “yes” and 13% said “no,” the reported level of integration between mill systems was significantly lower.
The method of integration between mill systems varied throughout the responses, as did the types of systems that were integrated. A few respondents indicated that all mill information systems were integrated, while one mill reported no integration at all, noting that “critical mill operations are segregated to avoid collisions and interference from the LAN.” Here are some representative responses from mills with information systems that share data:
• “Warehouse management/MES software and the profile management system share data for quality reporting.”
• “Data is shared on the mini-computer and via ODBC (open database connectivity).”
• “Maintenance data is on a separate Oracle database. Process data is captured and stored in a plantwide database.”
• “One system feeds other systems and is used to derive results. Data is pulled together from multiple databases for reporting.”
In addition, survey participants were asked how data was input between systems that rely on data from each other. Sixteen percent indicated automatically, 10% indicated manually, and 31% indicated both automatically and manually. Again, this year’s responses showed a less advanced level of connectivity than last year, when 33% indicated an automatic connection, 8% a manual connection, and 38% used both methods.
In this year’s survey, participants were also specifically asked if their process control systems in any way provide data to their information systems. Sixty-nine percent said that they did, while 23% said that they did not. Data sharing was accomplished either through software and network links or through direct hardware interface, as these comments demonstrate:
• “DCS and various other systems upload to our PI server for analysis.”
• “DCS sends data to MOPS [warehouse management system] via a chip link.”
• “The paper machine control computers share information with the finishing/shipping department.”
• “Data is provided to information systems through DDE [dynamic data exchange].”
• “Data is transferred via interfaces.”
• “A chip interface from Provox provides data to OSI’s PI.”
• “Data is passed via PI archive to other applications for status information.”
As for data sharing between mill information systems and corporate information systems, the majority of survey respondents described at least some level of integration (69%). Similar to last year, the majority of data sharing between corporate and mill systems was reported as either automatic (42%) or some combination of automatic and manual (42%). Here are some representative comments about this data sharing:
• “Some interfaces have been written, but because the enterprise’s applications are not open, some manual re-entry is necessary.”
• “Some data passes through an interface between Peoplesoft (ERP) and Maximo (CMMS). Also, data is entered by hand.”
• “It is accomplished through shared databases.”
• “Corporate sales system is fed by our production management information system.”
Previous experiences. Survey participants were asked to identify and describe any mill information systems that were second or third generation. Forty-four percent reported having at least one second or third generation system, which is in line with last year’s report of 48%. This year’s most commonly listed legacy system was the CMMS rather than administrative software. Other common answers were ERP, accounting, and warehouse management systems. Some mills reported that all their systems were second or third generation.
When asked whether they had been pleased or displeased with the older information systems, the majority of survey participants with such systems indicated at least some level of displeasure. The most common complaints were that the systems were slow and hard to maintain and administer. Other complaints included lack of functionality and user friendliness, as well as dependence on obsolete hardware.
In addition, survey respondents were asked to identify reasons for upgrading these second or third generation systems. The most common reason noted was that the old system was not Y2K compliant (22%)—an answer that did not appear in the 1998 survey. Other answers to the question varied and included the need to better serve customers, limited functionality of the old system, slow system speed, corporate or division IS strategy, the need to reduce manpower, poor integration, and the desire to take advantage of new technology.
Challenges. As with the 1998 survey (40%), the biggest challenge with installing and starting up new information systems was reported as user training and acceptance (31%). In this year’s survey, the second most prevalent answer was technological problems (21%), which was up from 13% the previous year. The reported technological challenges included data integrity, converting old programs to a new platform, programming, integration, designing applications that meet users needs, and debugging.
Obtaining money and management commitment for new information systems was also a commonly mentioned challenge (10%), as was procuring manpower resources for starting up the new system (7%). Other challenges included meeting project deadlines, business process re-engineering, and software licensing issues.
PERSONNEL AND TRAINING ISSUES. Participants were asked questions about the familiarity of personnel with certain aspects of computer hardware and software associated with information systems. Par- ticipants were also asked about mill training methods. Familiarity with hardware and software. When asked whether their mill workforce was more familiar with PCs or with mainframe terminals, 57% of respondents indicated PCs, while 5% listed mainframes, and 36% indicated both. In last year’s survey, 60% indicated PCs, while 10% listed mainframes, and 27% indicated both. This shift from either PCs or mainframes alone into the “both” category apparently shows the growing number of mill employees who have been trained to work on variety of platforms as software requirements have evolved.
In addition to asking about hardware use, the survey asked about the workforce’s familiarity with graphical user interfaces (GUIs) versus control-key, menu-driven software—the type most often associated with older mainframe systems. Responses demonstrated that the majority of users were more accustomed to both GUIs and menu-driven software (52%) rather than GUIs alone (16%) or menu-driven software alone (23%). The responses in the “both” category rose 6% from last year’s survey, again showing the growing knowledge of mill personnel.
Mill training methods. Compared with last year’s results (29%), the 1999 survey (46%) shows a growing tendency for mills to use a combination of both formal training and a “train-the-trainers” approach when upgrading information systems. This trend is not surprising, given that user training was listed as the biggest challenge for installing and starting up a system. Less participants this year (31%) reported that they would use only a “train-the-trainers” approach than in the 1998 survey (44%), while the use of formal training alone was about the same for both years.
BUSINESS DECISIONS. In addition to asking technology and personnel questions relating to mill information systems, Pulp & Paper asked questions about the business issues surrounding these systems. Respondents were asked about system cost and business drivers that have encouraged implementation of information systems, and mill satisfaction with the systems from an investment standpoint. The survey also asked mills about process, business, and supply chain impacts from information technology.
New systems. Thirty-four percent of mills responding to the survey indicated that they had either new or recently installed information systems, which was down 8 points from last year’s survey. The most common system listed was a CMMS (18%). ERP, order fulfillment, production tracking, accounting, and financial systems were also mentioned. In addition, respondents listed more networking and hardware upgrades than last year, which appears related to the number of smaller mills that responded. IBM AS/400 additions and LAN/WAN improvements were common answers.
Respondents were also asked about the costs for these new information systems, and were requested to include the costs for hardware, software, and any soft costs such as training, data conversion, and application interfaces. Total system costs ranged from $15,000 to $80 million, with a median expenditure of $377, 500.
In addition, survey participants were asked whether they would oversee computer upgrades at their mills, use a system integrator, or use a combination of both. The tendency to oversee such a project without the help of a systems integrator dropped from 56% to 41%. However, so did the tendency to rely solely on an integrator, with only 3% indicating this approach as opposed to 12% last year. Using both was the most common response, rising from 15% last year to 39% in 1999.
Ongoing computer support costs. Mills were asked if their ongoing computer support costs had increased with the use of new information systems. Unlike responses from last year’s survey, where 42% indicated that there had been an increase and 29% said there had not, 46% reported no increase and 39% saw some support cost increases in the 1999 survey.
Despite more mills reporting no increase, a high percentage still reported that they had seen rising computer support costs. However, the average increase reported (18%) was less than the average last year (25%). In 1999, responses ranged from 1% to more than 100%.
Investment satisfaction. When asked whether they were pleased with a new or recently installed information system, this year’s respondents were slightly less enthusiastic than last year (Figure 3). Less respondents answered “yes” (57%) than in 1998 (63%), while more indicated that they “don’t know yet” (41% in 1999, 37% in 1998). Some respondents were actually displeased with their new system this year, as opposed to none in 1998. The following comments are representative responses from mills with new or recently installed information systems:
• “The system works well, but needs better reporting.”
• “Yes, it has a faster response, is more fault tolerant, and provides better quality information.”
• “Yes, it is faster and more open to interfacing with customer shops.”
• “No, it is slow and cumbersome.”
Business drivers. As with last year’s survey, one of the most common business drivers leading mills to adopt new information technology was the need to integrate more with customers and provide better customer service (18%). One respondent specifically described the implementation of such technology as being driven by “the customers’ new technology of E-commerce and JIT product.”
Tying customer service as the most common driver for applying new information technology was cost reduction (18%). In addition, Y2K compliance was also commonly reported (15%)—an infrequent answer in the 1998 survey. The need to stay competitive, environmental compliance, layoffs and downsizing, and improved productivity were also cited.
Process impacts. When asked what process changes new information systems might provide, survey respondents reported a wide variety of improvements. The faster flow of better data was the most common answer, followed by improved monitoring, analysis, and feedback.
A more consistent product and better customer service, cost control, inventory control, reporting, customer service, and operator performance were also mentioned as process impacts, along with the removal of redundant tasks. The following comments are representative of the process changes noted by survey respondents:
• “It provides production and sales with better service to our customer base.”
• “We can spend more time analyzing data and less time managing data.”
• “We have faster, more complete information, and less process losses.”
• “Customer partnerships have become more of a reality.”
• “Better baseline monitoring, better trending, and easier troubleshooting.”
• “Operators know how they are performing, and they know right away if quality is good.”
Changes in business practices. As Figure 4 shows, those responding to the 1999 survey appeared unsure about the impact of information technology on business practices. When asked if they envisioned this technology changing such practices, less participants (49%) answered “yes” than in 1998, when 63% responded positively. In addition, the “not yet” category grew from 3% to 30%, showing a much more guarded outlook about the effects of implementing information systems.
Of the 49% who responded positively, improved customer service (25%) was the most common impact cited, followed by reduced paperwork and more productive employees. The following responses reflect attitudes about the positive impacts of information technology on business practices:
• “Customers, via the Internet, will be able to access order information, quality statistics, etc.”
• “We will share process and production information with customers.”
• “Streamlining and shifting emphasis from paper pushing to analytical functions will occur.”
• “Information technology will eventually connect quality-related issues with real-time process data.”
Supply chain insight. When asked whether new information systems at their mills had provided better insight into supply chain management, most survey participants answered “not yet” (48%) or “no” (38%), showing a much more “wait-and-see” attitude than was indicated in last year’s survey. Only 14% reported that supply chain management knowledge had improved.
In 1998, most participants answered “yes” (47%), while “not yet” received 13% and “no” received 40%. The cautious outlook in 1999 is probably related to the lower levels of integration reported between systems, as well as less ERP systems and less information systems overall. According to Tava Technologies senior vice president Tom Bruhn, some of this cautiousness is because paper companies have been “burned” in the past by this technology—for example, ERP.
“A lot of leading paper companies implemented ERP expecting tangible financial returns,” Bruhn explains. “They were also looking for business process changes, but ERP did not necessarily provide new functionality.”
“Paper companies are notoriously ‘make-to-stock’, and could benefit from a more synchronous management concept, which requires functionality not present in most ERP systems. This has possibly led to the paper industry’s conservative attitude toward new technology.”
Despite the more reserved attitudes reflected in survey responses, some participants with more advanced software implementations felt that they now had better business information. For example, one large fine papers mill with an extensive set of information systems described supply chain insight as being “better due to increased user-friendliness and improved interconnectivity with diverse systems.”
Such interconnectivity can be accomplished in various ways. For example, the use of data warehousing and decision support systems can provide a repository of data from various control and information systems throughout a mill or corporation. According to Deloitte Consulting/ICS partner G. Guy Sorrell, 46% of responding manufacturers in his company’s 1998 global survey of chief information executives indicated that they plan to use such systems within two years.
Sorrell sees such repository systems as crucial for companies trying to make the most of effective use of their information systems.
“Information access can provide strategic advantage to a company, but access alone fails to realize the full potential from technology investments,” Sorrell describes. “Application of emerging data warehousing technologies and related tools will allow harvesting of technology investments through improved knowledge sharing.”
| Web-Enabled Systems: The Next Step? |
| A small, but growing, number of software providers offer services over the Internet for a monthly fee, allowing users to avoid upgrade costs and licensing frustrations while providing access to the latest applications. For some companies, such services will make it easier to finance software purchases and maintain systems.
According to Tom Bruhn, senior vice president and general manager of Tava Consulting, a manufacturing consulting and systems integration firm, advanced uses of the Internet may be the wave of the future, despite some remaining reliability and performance issues that are associated with the Internet.
“Many of the leading software companies have conceded that client/server technology has matured and is now in the waning stages of its lifecycle, with ‘web-enabled’ technologies rapidly gaining momentum,” says Bruhn. “One of the biggest advantages of web-enabled and browser technologies is reduced network bandwidth requirements within a company.”
“When using client/server, you go to the server and serve the entire application over the network. You need sophisticated clients and networks for this, and they are very expensive to maintain. With web-enabled systems, all you usually need is the browser—everything else stays on the server you have logged onto.”
One of the growing number of companies that provide software services through the Internet is optAmaze. As its president, John Rachlin, explains, using the company’s paper machine scheduling and trim optimization services only requires an Internet connection and a standard web browser.
“A scheduler logs onto our web site and submits a problem through a secure channel,” describes Rachlin. “Problems are next relayed to a bank of high-speed, Java-based solvers. Results are then displayed to the user for analysis.”
|
| Open Systems Blur Software Category Boundaries |
| In the 1998 and 1999 Pulp & Paper information systems surveys, the request for mills to list software systems—CMMS, APS, MES, process control, WMS, and ERP systems—often yielded some confusing responses. In many cases, software applications were placed in categories that seemed to contradict the functionality of the product.
At first glance, confusion seemed to exist about the actual definition of the systems. However, a closer look revealed that open system technology, which allows simple data sharing between products, is a major reason that functional boundaries of products are blurring, making them difficult to categorize.
In this year’s survey, almost 40% or respondents reported some level of integration between systems. The number rose to almost 70% when respondents were asked if process data was shared with information systems.
To Mike Demaline, director of information products for Honeywell-Measurex, the most significant cause of confusion surrounding APS, MES, and WMS results from the fact that an integrated pulp and paper mill often contains batch, continuous, and discrete operations within a single manufacturing complex. He notes that APS, MES, and WMS products are designed either for batch industries, for continuous manufacturing industries, or for discrete manufacturing industries—but not all three.
“In the paper industry, systems that effectively solve customer-specific problems must often contain elements of APS, MES, and WMS functionality—all delivered as a fully integrated solution,” describes Demaline. “To achieve success, significant configuration, customization, and integration is required. The good news is that technology and open standards have evolved to the point where the integration can be nearly seamless.”
However, Demaline notes that it is the integration that has blurred the functional definition and boundaries of systems, causing confusion. “When the integration and customization are performed properly, it becomes difficult to determine where one system stops and another begins,” says Demaline.
Carol Jackson, pulp and paper industry manager for OSI Software, agrees that such system openness is crucial to meeting user needs, and that the software integration does “diminish” the functional boundaries of software packages.
“Most major software suppliers now publish APIs [application programming interfaces] or SDKs [software development kits] for programming access to their databases,” describes Jackson. “Others have opened up access with ODBC [open database connectivity], DDE [dynamic data exchange], or OPC [OLE (object linking and embedding) for process control]. Also, most leading software solutions incorporate a Microsoft client user-friendly interface. Users now find many packages easy to use with little or no formal training.They can perform simple integration among packages with a mere mouse ‘drag and drop.’”
“Barriers among systems are falling down, and paper companies are free to purchase 'best of breed' packages without having to worry about how these systems can exchange information, because integration is simpler than ever before.”
In Pulp & Paper’s 1998 and 1999 information system surveys, OSI’s PI package was mentioned in several categories. To Jackson, this is not necessarily surprising, given the open flow of data allowed by the PI system.
“PI was mentioned by some as an MES and by others as process control software, but it serves neither of these functions directly,” describes Jackson. “However, it does open the flow of information exchange among all of these systems using open integration techniques.”
“Users have created applications within the framework of their PIsystem that allow its functionality to grow outside of its original project scope. Data is historized in the PI system for use throughout the mill, and some mills have used its history to build advanced control applications, while other mills summarize production information and automatically forward it to cost accounting systems. So, to the engineer, PI might be a process control tool, while to the accountant, it might be a source of real-time grade costs.”
Another system placed in a wide variety of software categories was EMPAC (enterprise maintenance planning and control) from Indus International. EMPAC is an advanced CMMS that features total integration of stores, purchasing, and work order management. It also features an asset optimization module that accesses data from PLCs.
To John Yolton, senior industry consultant for Indus, EMPAC’s access of PLC data might be one reason for its misleading categorization as process control software.
“Through open architecture, the asset optimization module collects information about a piece of equipment from its process control devices,” describes Yolton. “A diagnostics tool in EMPAC informs the user of where the problem is and then suggests a maintenance or operating response.”
While sometimes confusing, Yolton, like Demaline and Jackson, ultimately see this blurring of functionality as a great benefit for the user.
“ERP companies like Oracle and Peoplesoft know that customers don’t necessarily want to forego functionality, so they partner with ‘best-of-breed’ companies like us,” says Yolton. “With open systems, customers don’t have to accept ‘diluted’ functionality in software packages.”
|
|
