The “Gentler” Giant’s Value Proposition to Overcome the “Little Guys’” Perception

The “Gentler” Giant’s Value Proposition to Overcome the “Little Guys’” Perception

Software giant Oracle, a long-time provider of enterprise software systems for the largest of corporations worldwide, has now set its sights on the small to medium business (SMB) market, thereby giving Microsoft and other leaders in this segment the kind of competition they have never had to contend with before. The vendor has begun a campaign to attract smaller partners in an effort to show SMBs that its offerings are viable choices for their organizations. For more background, please see part one of this series, A “Gentler” Giant’s Success in Reaching Out to the “Little Guys.”

Strong Product and Capabilities Still Matter

As mentioned in part one of this series, resellers have to work directly with value-added distributors (VADs) that are Oracle Remarketer Authorized to transact orders using Oracle's standard terms, conditions, and pricing for Oracle's 1-Click Ordering Programs, which currently include the following:

* Oracle Database Standard Edition and Standard Edition One, including Oracle Warehouse Builder, an extract, transform, and load (ETL) tool, and Oracle Application Express, a rapid development tool for Web applications atop the Oracle database.

* Oracle Application Server Standard Edition and Standard Edition One, and Oracle Application Server Java Edition—application servers that are all part of the Oracle Fusion Middleware family for hosting a company's web site and applications. This middleware suite also provides an instant portal for creating an extranet and content management for managing unstructured information, such as purchase orders, marketing collateral, or presentations. Oracle Business Activity Monitoring (BAM) is an additional tool.

* Oracle Business Intelligence Standard Edition One—this business intelligence (BI) product provides users with access to dashboards (via Oracle Interactive Dashboards) to format and distribute reports (via Oracle BI Publisher) and to enable ad hoc analysis of data integrated from disparate sources (via Oracle Answers).

Oracle Database is the company's flagship product and the main element of the Oracle VAD Remarketer Program discussed in A “Gentler” Giant’s Success in Reaching Out to the “Little Guys.” . In mid-2007, the vendor launched Oracle Database 11g. This product, which at this stage is not part of this initiative, features new technology to accelerate the adoption of database grids, improve storage, and simplify access to data in online analytical processing (OLAP) cubes.

The previous version of the company's flagship database, Oracle 10g R2, came out in 2005, two years after the original version of Oracle Database 10g. The "g" in 10g and 11g stands for "grid," as in grid (or utility) computing. With 10g, Oracle introduced the feature that it refers to as Real Application Clusters (RACs). These are a way to join servers running Oracle's database together to work on database queries in parallel (see Oracle Further Orchestrates Its SOA Forays). According to Oracle, about half of its database customers have upgraded to 10g, with a fraction using RAC grids.

Among 11g features that Oracle hopes will attract more users to its grid computing systems are Oracle Real Application Testing and Oracle Data Guard. These features allow the splitting of grids to permit the testing of upgrades or system changes before moving them into production, as well as to facilitate backup and disaster recovery. Storage improvements in the latest release include automated data partitioning, better data compression, a feature for backing out of delinquent transactions, and the Oracle Total Recall capability. Oracle Total Recall allows administrators to run queries against the data as it stood at a specified point in the past. Finally, Oracle Database 11g allows native integration with the Microsoft Visual Studio 2005 development tool set, which was done with the hope of attracting Microsoft partners.

Until recently, Oracle’s research and development (R&D) emphasis has been on performance and scalability (i.e., customer satisfaction and product quality) rather than price and ease-of-use for the masses. The company remains the overall database market share leader, and it now insists that, after years of playing and trying, it has finally gotten the ease-of-use (employee productivity) and packaging formula (i.e., a better solution value—more features for lower cost) established, and it is ready to make a renewed push to small businesses. In fact, Oracle touts many recent benchmarks that purport its leadership in database performance and value, such as that Oracle Database 10g is 45 percent faster and 14 percent cheaper than Microsoft SQL Server 2005.

According to the Transaction Processing Performance Council (www.tpc.org), the Oracle database is also the price and performance leader in the TPC-C online transaction processing (OLTP) performance benchmark and the TPC-H data warehouse performance benchmark. Also, Oracle Database 10g Standard Edition One comes cheaper than Microsoft SQL Server 2005 SE, in terms of both the number of users and the number of sockets. Given that the database is also tightly integrated with Microsoft Windows, Oracle also touts a better (i.e., lesser) price and performance figure on that operating system too.

The typical problem with non-Oracle databases is when a small customer calls and asks the resellers for a small server to build a database. The straightforward solution is to sell the customer a two-way server with a standard Microsoft SQL Server database. Six months later, the same customer might call again and ask for a little more horsepower, in which case the reseller will offer a four-way server and more standard SQL Server software. However, some time down the road, when the customer’s business is really booming and it could use even more horsepower, a dead end appears in terms of a more scalable offering.

Conversely, the Oracle alternative is to sell this customer a two-way server from Hewlett-Packard (HP) with Oracle Database 10g Standard Edition One. Several months later, that solution is still working great, but the company could use a little more horsepower. The Oracle reseller now has the following options:

Managing the Overflow of E-mails

The impact of e-mail on businesses these days is enormous. Companies use e-mail to conduct business, for responding to clients, for internal communication, for discussing strategy, and for responding to regulations. Roughly 35 billion e-mails are sent a day, and the number is growing. The Radicati Group estimates that by the end of 2006, 52 billion e-mails will be sent daily, including all types of correspondence. More importantly, 60 percent of business-critical data is sent via e-mail, according to Gartner.

E-mail Evolution

Unsupervised and improperly disciplined e-mail behavior causes headaches to corporate management. Viruses hiding within e-mails can cause huge harm to organizations. The fact that so many e-mails are sent every day means that there is a chance of inappropriate e-mails (with respect to virus content, sensitivity or privacy issues, incorrect addresses, and so on) slipping through. Yet organizations do need to focus on storing business-critical information, and it is essential that this information is properly stored so that it is accessible and reusable.

E-mail management aims at the preservation of e-mails and the information contained within them. Historically, e-mail management consisted only of storing and preserving e-mails to optimize server efficiency. The focus of e-mail management today, however, has shifted to addressing regulatory compliance more than anything else.

The current market seeks to integrate e-mail management with a full document and records management (RM) solution. This enables organizations to index e-mails, and provides users the ability to search and to use the repository as a knowledge archive.

Currently, most content management vendors have already moved towards integrating an e-mail management solution with an RM solution. Examples include EMC/Documentum's acquisition of Legato, a provider of e-mail archiving products, and IBM's integration with iLumen, a provider of enterprise message management tools.

Knowing What to Store

Storing e-mails or any other kind of data is not problematic for organizations. The cost of extra storage space has decreased significantly over the last decade. The problem is knowing which e-mail content to store and preserve, and which to destroy.

Organizations need to set forth policies on e-mail usage. Policies should address not only e-mail use and misuse, but also retention and destruction. According to research conducted by the Association for Information and Image Management (AIIM), 80 percent of organizations have some kind of policy for e-mail use, but 60 percent have no formal policy governing its retention. Organizations need to rethink their strategies on e-mail management: currently, 31 percent of all organizations keep e-mails indefinitely, and preserve 26 percent of e-mails for less than 120 days, or establish a maximum storage on people's inboxes to limit the retention.

The problem with this approach is that inadequate restrictions or improper methods can eventually harm the organization itself. For example, internal e-mails regarding lunch appointments do not need to be saved, while an e-mail regarding a lunch appointment with a business client should be. Saving all e-mails in the hopes of preventing some information loss is not the right solution. On average, according to AIIM, 75 percent of e-mails are not useful for saving, but finding the 25 percent that are important can be quite difficult.

One potential solution is to create business rules within archiving software. These business rules avoid unnecessary storage, as users can flag the e-mails they want to archive. Defining these rules has to be a joint effort between the information technology (IT) department and the business sector to make sure it covers more than just the technical side of actual e-mail storage. Flagging should be made possible based on a company name, keywords, subject or message text, the sender, or even the software used to send the e-mail.

However, users should not have full control on what should be saved within the e-mail management business rules: this should be controlled at a systems level. This way, the basic principles will be governed at the administrative level, while individuals can refine the rules on a more personal level.

Manage Volume and Risk

The volume of e-mails grows daily. And it's not just the number of messages, but also their size. Organizations deal with all sorts of messages: chain letters, joke-of-the-day e-mails, lunch meetings and arrangements, business e-mails, and so on. E-mail also grows in size because of attachments, pictures, movies, large documents, and even colorful signatures with company logos or theoretically witty sayings. Also, the misuse (or lack of knowledge) on the part of undisciplined users leads to unnecessary duplication of e-mails through copying and forwarding and replying, which in turn causes capacity overload. Because of this overload, organizations have to focus on managing e-mail volume.

Even though storage decreases in price every year (a 550-gigabyte hard drive currently sells competitively for about $400 [USD]—that is, $0.80 [USD] per gigabyte), retaining too many e-mails means storing needless data. The cost of retrieving the data (and the risk of not being able to find the right information within an appropriate time frame) is even more important than just the cost of storage.

Can Enterprise Applications Meet the Challenge?

Nowadays manufacturers are increasingly subject to massive pressures due to the need for driving down costs and increasing efficiency. What makes things worse is that with product life cycles decreasing, manufacturing and distribution are increasing in complexity. This, for the manufacturer, translates into a need to better manage customer demands and expectations and to respond accordingly. Furthermore, manufacturers of electrical and electronics equipment must comply with a growing array of strict environmental regulations, many of which have already been implemented in the European Union (EU) and the United States (US). More regulations are pending in Japan, China, and other countries. As in many other industries, the cost of compliance can be high, but the cost of noncompliance can be far greater. Thus, the industry winners have to gain the capabilities they need to adapt their businesses to meet regulatory requirements—from product design to compliance reporting, and from sourcing and procurement to service and repair—so that they can avoid costly penalties and product recalls, optimize processes to comply with changing regulations, build trusted brands, and protect shareholder value.

Such manufacturers will have to turn somewhere to comply with these high-tech and electronic industries' significant and stringent environmental policies. Specialized, private marketplace service providers that offer auction platforms to off-load a company's excess and obsolete (E&O) inventory are the logical outlets for manufacturers to use in order to ensure compliance with these new regulations. Ideally, these providers should have an established number of treatment recycling and transportation management company partnerships. An environmental policy came into effect in August of 2005 for member states of the EU. The Waste Electrical and Electronics Equipment (WEEE) Directive 2006/96/EC sets recycling and reuse standards across a variety of industries from home appliances to computer products. The WEEE directive holds the manufacturer (producer) ultimately accountable for recovering products and for recycling up to 75 percent of the material content by weight. Failure to comply results in the manufacturer paying a penalty of 2 percent of its annual revenue. In other words, the WEEE directive establishes rules for the collection, treatment, recycling, and recovery of electronic waste in the EU. It states that electronics manufacturers and importers must manage and pay for the recycling of electrical and electronics waste.

In addition, the WEEE legislation's directive states that electronic product manufacturers, excluding retailers and distributors, are responsible for providing take-back programs for all electrical and electronic equipment sold in the EU's member states, as well as in Norway and Switzerland. The directive defines, prescribes actions, and sets regulatory milestones for the collection, treatment, recovery, and financing of discarded electrical and electronic equipment across ten product categories. These ten categories range from information technology (IT) and telecommunications equipment, large and small appliances, and tools to toys and leisure equipment. Naturally, product reuse (that is, the resale or reuse of whole appliances for their original intended function) is to be given priority over recycling. For IT equipment, telecommunications, and consumer electronics that do not have a whole product reuse option, 75 percent of the product weight must be proven to be recycled. New products must be marked with "do not trash" symbols, and information on product disassembly must be provided by manufacturers. The target date for commencement of these programs was August 13, 2005. Since then, the EU member states have been obliged to provide for the financing of the collection, treatment, recovery, and environmentally sound disposal of waste electrical and electronic equipment. They have had to set up separate collection systems to eliminate the disposal of such products into municipal waste. To that end, distributors must ensure that waste of the electronics equipment can be returned to them free of charge, and manufacturers must set up and operate individual or collective take-back systems.

Since December 31, 2006, EU member countries have had to meet WEEE recycling targets in that the rate of recovery for IT, telecommunications, and consumer equipment is at least 75 percent, which is measured in terms of average weight. Manufacturers must now state the weight of the electrical and electronic waste entering and leaving treatment and recovery or recycling facilities. Member states must draw up a register of manufacturers along with the quantities and categories of electrical and electronic equipment placed on the market, collected, recycled, and recovered in their territory. Each member state must also transpose the WEEE legislation into local law, which is where local differences create WEEE compliance reporting issues even though there is general adherence to the EU level directive. The task of monitoring manufacturers' sales in volumes to each country (for the purpose of establishing recycling quotas) will fall to a member state's agency working under the direction of its national Office of the Environment as the managing authority for WEEE. On their side, manufacturers must register up front with each country's authority for the purpose of reporting recovery and recycling results. The initial recycle quota is set at a relatively low bar of 4 kilograms per capita per year, although countries such as the Netherlands have had established programs that exceed this volume for years.

Although the WEEE directive has jurisdiction only over the EU, most multinational electronics and telecommunications companies will implement the infrastructure and IT necessary to manage compliance processes on a global basis. They do this in anticipation of similar legislation in other regions and to maintain worldwide process standardization. With legislation like WEEE, supply chain management (SCM) and product lifecycle management (PLM) have become cradle-to-grave endeavors with significant depth and complexity added to the reverse logistics process. But an even bigger burden might be the requirement for manufacturers to recycle a portion of electrical and electronic waste made way back when, which seems a daunting task, and the specifics of how it will work exactly are still largely unknown.

Advancing the Art of Pricing with Science

Though companies recognize the need for a better way to manage their pricing strategies, many continue to lose money by using archaic pricing methods. But there is a new approach beginning to surface in the market of price management. Science-based software can be leveraged to help companies create more accurate and complete pricing strategies in order to meet their margins. To learn more, please see Know Thy Market Segment's Price Response.

In 2005, Zilliant, an Austin, Texas (US)-based provider of data-driven, strategic pricing applications, and the Institute for International Research (IIR) released the results of a survey that showed strategic pricing was gaining in priority among some US businesses. The PriceX Conference poll surveyed nearly seventy businesspeople responsible for making pricing decisions at their respective companies.

Despite this finding, adoption of strategic, science-based pricing and associated technologies is relatively minor in industries other than airlines, hotels, and retailers. These industries practice a form of science-based pricing called yield management. Yield management, also known as revenue management, was invented three decades ago; its goal is to fill as many seats and rooms as possible while charging the highest prices the market can bear. Since then, these industries have adopted sophisticated software programs to predict demand and to set prices, resulting in as many different price points per flight as passengers, or per room as guests.

Armed with a wealth of customer data, programmers then developed formulas that could manipulate prices up or down depending on existing sales, the likelihood of last-minute purchases, and other variables ranging from weather forecasts to competitors' prices. The underlying logic was that airplane seats and hotel rooms are worthless if unused, and selling them even at a loss meant gaining some revenue.

Given that "computer power" is much more affordable these days, user enterprises can harness statistical science to analyze transactions and other customer data to more accurately explore the cause-and-effect relationship between prices and purchase decisions. The idea here is to be able to discern customers' "willingness to pay," and set "take-it-or-leave it" prices where companies will make the maximum revenue.

Using mathematical formulas and massive databases of sales records, companies can forecast their sales plans, and test pricing and demand elasticity under various discount or package scenarios before trying them in the market. Layering in data from other customer interactions can help companies set prices, schedule markdowns, and identify top performing buyers with more sophistication than ever before. Companies can also set prices based on the value consumers derive from specific products, or even plan different discounting and pricing strategies based on anticipated customer behavior.

Again, as indicated earlier on (see Know Thy Market Segment's Price Response), business-to-business (B2B) pricing environments are different in that pricing is opaque and largely discretionary. Now that technologies have been brought to market that address these dynamics, B2B companies are getting on board too.

Nevertheless, according to Zilliant, although pricing is generally accepted as a core business practice, the process most B2B companies go through in determining a price is often archaic and arbitrary. Some businesses simply take the cost of a product and add margin on top of that price, while others simply match or better their competitor's offering. Another common practice is the so-called "out of thin air" (OTA) or "sucking (knowledge) out of my thumb" method; in other words—guessing. According to the above mentioned PriceX survey, 56 percent of companies polled have some sort of pricing strategy in place, while only 44 percent have a dedicated pricing department or an individual with pricing responsibility.

Other key trends uncovered in the survey included that 35 percent of companies consider pricing to be a top priority, yet 61 percent of companies use Excel spreadsheets to determine price (rather than specialized pricing applications from a vendor). Data cleansing was cited as the main obstacle to improving pricing policies, followed by ineffective customer segmentation.

In a somewhat older survey (taken a few years ago) by the Professional Pricing Society of its members, 30 percent of respondents said they priced new products by mirroring their nearest competitors' prices, and another 22 percent set prices for new products based on recovery of costs and to tack on a profit. Only 18 percent revealed that they performed some sort of customer research to determine the value of the product or service to potential customers. And when it comes to the Internet pricing, 40 percent said they simply mimic the pricing of their off-line sales channels, and 28 percent responded that they do not have an Internet strategy at all.

In other words, most businesses lack a detailed understanding of their market segments' responses to prices and deal terms. They rely solely on undifferentiated discount policies and sales team discretion to structure all types of deals, from quotes to orders, agreements to contracts. As a result, some deals go through with overly generous terms, while others are lost due to gross pricing misalignment.

These harmful practices continue to take place despite some pricing pundits "shouting blue murder" (protesting) about the ingrained, casual thinking that pervades the global economy regarding pricing. Both consumers and businesspeople erroneously assume that price has everything to do with cost. Yet, while any company has to know the cost of a product, it is only so that it can understand the profitability implications of the price, not for the purpose of setting the price. The value (benefit per unit price) is in the eye of the customer and depends on the circumstances surrounding the deal. Another faulty practice is the assumption that when a company is in a competitive situation and prices drop, the company must match the price-drop. Also, executives who are devoted to using data and analytics in all kinds of other functional areas still think it is entirely acceptable to set prices based on "history," "experience," or "instinct."

The Necessity of Data Warehousing

Data warehousing is an integral part of the "information age". Corporations have long known that some of the keys to their future success could be gleaned from their existing data, both current and historical. Until approximately 1990, many factors made it difficult, if not impossible, to extract this data and turn it into useful information. Some examples:

* Data storage peripherals such as DASD (Direct Access Storage Device) were extremely expensive on a per-megabyte basis. Therefore, much of the needed data was stored offline, typically on magnetic tape.

* Processing power was very expensive as measured in MIPS (Millions of Instructions per Second). Mainframes had to reserve most of their processing power for day-to-day operations, reports could only be run overnight in batch mode (without interaction from the user).

* Relational database technology was still in its infancy, and server engines were not powerful enough to support the data loads required.

* The type of programming that had to be done with third generation languages (3GL's) was tedious and expensive. Fourth generation languages were needed to abstract some of the required coding, but 4GL's were still in their infancy.

Most operational data is stored in what is referred to as an OLTP (On-Line Transaction Processing) system. These systems are specifically designed for high levels of transaction volume with many concurrent users. If the database is relational, it has probably been "normalized" (the process of organizing data in accordance with the rules of a relational database). If the database is non-relational, custom programs have to be written to store and retrieve data from the database. (This is often accomplished with the COBOL programming language). Whether relational or non-relational, the very design that makes an OLTP system efficient for transaction processing makes it inefficient for end-user queries. In the 1980's, many business users referred to their mainframes as "the black hole", because all the information went into it, but none ever came back out - all requests for reports had to be programmed by the Information Systems staff. Only "pre-canned" reports could be generated on a scheduled basis, ad-hoc real-time querying was virtually impossible.

To resolve these issues, data warehousing was created. The theory was to create a database infrastructure that was always on-line, contained all the information from the OLTP systems, including historical data, but structured in such a way that it was fast and efficient for querying. The most common of these schemas (logical and physical database designs) is known as the star schema. A star schema consists of facts (actual business facts) and dimensions (ways of looking at the facts). One simple way to look at a star schema is that it is designed such that the maximum amount of information can be derived from the fewest number of table reads. Another way to reduce the amount of data being read is to pre-define aggregations (summaries of detail data, such as monthly total sales) within the star, since most queries ask questions like "how many were sold last month?"

Data warehousing also led to the development of the concept of metadata management. Metadata is data about data, such as table and column names, and datatypes. Managing metadata makes it possible to understand relationships between data elements and assists in the mapping of source to target fields. (For more information of Metadata see "Metadata Standards in the Marketplace ")

Next came the creation of Extract/Transform/Load (ETL) tools, which made use of the metadata to get the information from the source systems into the data warehouse.

Additional tools, which made use of SQL (Structured Query Language), were developed to give end-users direct access to the data in the warehouse. As time went by, the query tools became user-friendly, and many now have a parser that can turn plain English questions into valid SQL. These end-user tools are now loosely referred to as "business intelligence" tools. In addition, there are other database constructs used to assist business intelligence tools in multi-dimensional analysis of data in the warehouse. These databases are referred to as hypercubes (also known as cubes, multi-dimensional cubes, or MDB's).

Since the early 1990's, data warehouses have become ubiquitous, technology and methodology have been improving, and costs have been decreasing. In 1998, data warehousing was a $28 Billion (USD) industry, and growing at over 10% per year. In addition, a recent survey of top IT executives indicated that data warehousing would be the number one post-Y2K priority. Data warehousing is now recognized as an important way to add business value and improve return on investment, if it is properly planned and implemented.

Selection Issues

Selecting a set of products for a data warehouse effort is complex. The first and most important issue is to ensure that the Extract/Transform/Load tool that is chosen can effectively and efficiently extract the source data from all the required systems.

The selection of the ETL tool requires an understanding of the source data feeds. The following issues should be considered:

* Many warehouses are built from "legacy" systems that may be difficult to access from the computer network. ETL tools often do not reside on the same machine as the source data.

* The data structures of the legacy systems may be hard to decompose into raw data.

* Legacy data is often "dirty" (containing invalid data, or missing data). Care must be taken in the evaluation of the tool to ensure it has an adequate function library for cleansing the data. Depending on the complexity of the cleansing required, a separate tool designed specifically for cleansing and validation may have to be purchased in addition to the ETL tool.

* The ETL tool should have a metadata ("data about data") repository, which allows the data sources, targets, and transformations to be tracked in an effective manner.

* The tool should be able to access legacy data without the need for pre-processing (usually with COBOL programs) to get the data into sequential "flat files". This becomes increasingly complex when working with filesystems like VSAM (Virtual Sequential Access Method), and files that contain COBOL Occurs and Re-Defines clauses (repeating groups and conditionally defined fields). It should be noted that a large percentage of the world's data is stored in VSAM files.

* A final issue is whether the ETL tool moves all the data through its own engine on the way to the target, or can be a "proxy" and move the data directly from the source to the target.

Selection of the business intelligence tool(s) requires decisions such as:

* Will multi-dimensional analysis be necessary, or does the organization need only generalized queries? Not all warehouse implementations require sophisticated analysis techniques such as data mining (statistical analysis to discover trends in the data), data visualization (graphical display of query results), or multi-dimensional analysis (the so called "slice and dice").

* Will the architecture be two-tiered or three-tiered? Three-tiered architectures offload some of the processing to an "application server" which sits between the database server and the end-user.

* Will the tool employ a "push" or a "pull" technology? ("Push" technology publishes the queries to subscribed users, much like Pointcast works, "pull" requires that the user request the query).

* Will the information be broadcast over a corporate intranet, extranet, or the Internet?

* How will the organization implement data security, especially if information is being broadcast outside the corporate firewalls?

A One-stop Event for Business Intelligence and Data Warehousing Information

The Data Warehousing Institute (TDWI) hosts its quarterly World Conference in cities across the US to help organizations involved in data warehousing, business intelligence (BI), and performance management, by giving them access to industry experts, and providing impartial classes related to topics pertinent to the industry. As the industry grows, organizations are faced with questions about how to best access their data to drive profits and meet goals and budgets. The need to understand data warehousing and the best means of leveraging data has become essential to developing a forward-looking approach to a BI solution.

This year, TDWI's summer event was held in San Diego, California (US) from August 20 to 26. Participants were able to take advantage of courses given by worldwide BI experts, as well as network with peers, have access to vendors and product demonstrations, and participate in one-on-one sessions with industry experts and instructors. The six-day event provided one-stop shopping for participants, who were able to take advantage of planned networking events, a two-day trade show highlighting various vendor offerings, and classes ranging from data warehousing testing techniques to best practices in performance management. The advantage of this one-stop shopping approach was that organizations had the opportunity to evaluate software, compare vendor offerings, and gain knowledge from other organizations that have implemented their own data warehousing environments.

The conference focused on five main themes, namely business analytics, leadership and management, data analysis and design, data integration, and administration and technology. These themes identify the main areas within data warehousing and BI, and provide the necessary knowledge related to the whole design and implementation process. A series of classes were offered in each area to allow users to focus on a specific industry aspect, or to gain an overall understanding of the sector and the different driving forces within it. Not only does TDWI focus on technology and the drivers associated with technological advances, but a key advantage to participating in the conference series is the additional focus on the business side of technology, and on managing the business processes associated with BI and performance management.

TDWI Overview

TDWI delivers research, education, and news, which enables individuals, teams, and organizations to leverage BI industry information to improve organizational decision-making, optimize performance, and achieve business objectives. One of TDWI's goals is to provide organizations with the impartial information required to make informed decisions. Although the organization runs events sponsored by various vendors, and provides users with product-related information, TDWI touts itself as being a central impartial resource for information. Business and information technology (IT) evaluators of solutions—whether in the requirements-gathering or enhancement phase of a current platform—can access a wide range of information, including classes, webinars, on-site training, and research.

TDWI has an international membership program, and provides industry publications and news, and a comprehensive web site. A division of 1105 Media, TDWI was created in 1995. It has over 5,000 members from Fortune 1000 companies, and includes both business and technology professionals. It is regarded as one of the central organizations for collecting data and providing insight into the world of data warehousing and BI.

TDWI collects and promotes best practices research to educate technical and business professionals about new BI technologies, concepts, and the approaches that have been applied in other organizations. This research also addresses significant issues and problems that organizations have experienced, and how they were handled. Many companies use TDWI's information to identify how they measure up to industry standards, how to take advantage of new or upcoming technologies, and how to address issues that relate to how they conduct business. The benefit of this information is two-fold. First of all, organizations can keep on top of enhancements within the industry, and can gain a wider knowledge base than that provided to them by their service provider (their selected vendor). Secondly, TDWI can help drive industry trends by leveraging the needs of organizations, as well as the way vendors should develop products to meet those needs.

TDWI's annual BI Benchmark Report identifies best practice metrics and compares TDWI's data warehousing maturity model to the industry. Many organizations consult this report to benchmark their BI use to ensure they are optimizing their implemented solutions, and discover ways to continuously improve their technical platforms and BI environments. This can include comparing their current environment with other organizations, or looking at information about other organizations within their vertical markets. TDWI also distributes other industry-related publications:

* The Business Intelligence Journal, published biannually, provides information and resources for BI and data warehousing professionals. The focus is on actionable advice on how to plan, build, and deploy BI and data warehousing solutions.
* Ten Mistakes to Avoid is distributed quarterly, and advises readers on different topics related to building, deploying, or maintaining a data warehouse, or managing a data warehouse team.
* What Works: Best Practices in Business Intelligence and Data Warehousing, also distributed quarterly, gives readers a comprehensive collection of case studies, questions and answers, and lessons learned from the experts.
* TDWI e-mail newsletters provide up-to-date news and industry commentary.

These publications provide users with continual information on the industry, and can help identify pitfalls in order to prevent them from making those same mistakes. Also, organizations that are in the same situations can gain insights on how to solve issues, as well as learn from other organizations and industry experts.

TDWI also develops webinars to discuss pertinent issues in the BI and data warehousing industry, and gives training at customer sites. TDWI seminars deal with the skills and techniques used to ensure successful implementations of BI and data warehousing projects. Overall, TDWI leverages its decade of experience within the data warehousing and BI industries to provide organizations with the information needed to make the best decisions possible. This way, organizations can access information that is industry-specific (without a bias towards one vendor versus another), and benchmark their own BI and data warehousing environments against organizations that have more experience implementing and growing these solutions. Also, organizations can compare and contrast challenges and issues as they arise.

TDWI Conference Tracks

Each of the quarterly conferences focuses on different tracks. These tracks present business and IT users with classes and seminars that highlight main industry trends, and provide a basis for enhancing their current data warehousing and BI environments (or aid in the requirements and selection process to implement such an environment). Not only do classes provide a wealth of information that can be justly described as verging on information overload, but in-class exercises, depending on the class, allow users to internalize the information to which they are being exposed. Aside from diverse and in-depth topics, the instructors are experts (whether within their respective industries, or their consultancy practices). Not only can users learn about the topics being presented, but they can also meet with experts to gain additional insight into topics directed specifically to their organizations.

Over fifty classes were offered during this summer's six-day conference. Topics ranged from data warehousing testing techniques to performance management benchmarking practices, in either full-day or half-day sessions. This allowed participants to learn about the latest trends, best practices, and industry insights on how to improve their current structure or enhance their technical platforms. Five tracks were presented during the event:

* Business Analytics
The business analytics track focused on both business and technical aspects of analysis. Topics included performance management, the definition and delivery of business metrics, data visualization, and the deployment and use of technology solutions. Solutions discussed included online analytical processing (OLAP), dashboards, scorecards, and data mining, as well as analytic applications. This focus allows organizations to gain insight into areas within BI and the different aspects of insight that analytics can provide. Organizations that require a subset of BI can identify how their needs can be met, by identifying requirements based on the topics presented. Additionally, they can take advantage of the trade show to identify those vendors that meet their needs, or those that (while not all-encompassing BI vendors) play in a specific space within the industry, such as data mining or data integration.

* Leadership and Management
The leadership and management track provided users with the insights needed to take a project from inception through to completion. Aside from identifying process and project management methodologies related to data warehousing and BI projects, emphasis was placed on the overall management of these projects. Ideas presented ranged from team building and the high level technical requirements needed to manage such projects, to other business areas such as customer relationship management (CRM) and supply chain management (SCM). This focus allowed users to identify a broad range of topics and considerations needed to implement and manage a data warehousing project through the systems development life cycle. Additionally, outside markets were identified to show the interrelation between BI and other industries. For example, many operational BI efforts are driven by SCM and the need to manage day-to-day decisions from the shop floor.

* Data Analysis and Design
A key focus of the data analysis and design theme centered on the skills needed to identify business needs and to transform those needs into data structures that are adaptable, extensible, and sustainable to the business unit. Course topics included needs analysis, specifications of business metrics, and data modeling. These topics and surrounding concepts encompass the backbone of developing a data warehousing and BI platform. Identifying business and systems requirements and translating them into the appropriate systems requirements is essential within any project. Within data warehousing and BI, it becomes more important as platforms are designed, and as business needs analysis has to be integrated into the actual design of the platform. Integration questions center on whether the current systems will integrate with the new software—and more importantly, how they will integrate.

* Data Integration
The theme of data integration included all the topics related to implementing a data warehouse solution. Included were data profiling; data transformation; data cleansing; source and target mapping; data cleansing and transformation; and extract, transform, and load (ETL) development. It is important not to underestimate the importance of data integration, as the way data is integrated into a data warehouse or BI solution is the essence of that system. If a scorecard is developed to measure an organization's sales metrics and the source data is not accurate, the key performance indicators (KPIs) set and reported on will be meaningless.

* Administration and Technology
The administration and technology track identified and covered topics related to infrastructure management, and the continued successful operation of data warehousing and BI solutions. The focus was on technology architecture, planning and configuration, system and network administration, database administration, and access and security administration. Maintenance of the implemented architecture and platform is essential to continued success in the data warehousing and BI environment. This section helped bridge the gap between administration and technology, and identifies the complexity of managing these two aspects of a data warehouse.

Microsoft Goes Their Own Way with Data Warehousing Alliance 2000

"REDMOND, Wash., Nov. 30 /PRNewswire/ -- Microsoft Corp. (Nasdaq: MSFT) today announced that 47 applications and tools from 39 top vendors throughout the industry have qualified for Microsoft Data Warehousing Alliance 2000. Alliance members and partners are committed to delivering tools and applications based on the Microsoft Data Warehousing Framework 2000, an open architecture for building business intelligence and analytical applications based on the open standards and services built into the Windows 2000 operating system, Microsoft SQL Server 7.0 and Office 2000. Application vendor membership for the Data Warehousing Alliance has more than doubled since it was originally announced in October 1998."

According to the release "organizations leveraging the framework and using alliance member products are better able to align local decision-making around key business drivers and harness the full potential of the web to win new customers, retain and extend customer relationships, and work more effectively with partners."

The architecture is based on OLE DB and the Open Information Model (OIM), in "recognition of the value and competitive advantage provided by the data warehousing services built into Microsoft products."

According to Microsoft, this technology is based on the Microsoft Data Warehousing Framework, which "is based on open, published protocols for interoperability and integrated end-to-end data warehousing services. It utilizes technologies provided in Microsoft Office 2000 and Microsoft SQL Server 7.0 products, and a partnership with Data Warehousing Alliance members for complementary tools and applications. The DWF enables data warehousing solutions where the data comes from virtually any source and where any type of information can be delivered to any compliant client interface or application."

Market Impact

Once again, Microsoft is using proprietary standards (OLE DB and OIM) to achieve its data warehousing goals. The more widely accepted standards are under the stewardship of the Object Management Group (OMG), which has over 800 members. OIM is a standard developed by Microsoft and turned over to the MetaData Council (MDC) which has "close to 50" members. For more information on the dueling standards bodies see "Is There Finally A Metadata Exchange Standard on the Horizon?", (http://technologyevaluation.com/news_analysis/09-99/NA_DW_MFR_9_28_99_1.asp,September 28, 1999). The alliance criteria require compliance with OLE DB for data access and the Open Information Model for sharing metadata. According to Colin White, president of DataBase Associates International Inc., "The Microsoft Data Warehousing Framework 2000 makes it easy to build Digital Dashboard applications integrating business intelligence, collaboration, and Web content right into the environment many knowledge workers live in: Outlook 2000."

This effort should make it easier for customers to integrate and use tools from multiple vendors, as long as their database is Microsoft's, and the other vendors are members of this alliance. The web component is to be provided by Microsoft's SQL Server 7.0, a component of the Windows DNA platform (Distributed interNet Architecture, introduced in 1997, Microsoft's umbrella term for its enterprise network architecture based on COM and Windows 2000 (NT 5.0)). The Windows DNA platform is advertised as "Microsoft's comprehensive platform for building Web applications."

We believe this will only serve to further fragment the data warehousing market. Obviously, Oracle is not a member of this alliance, and other applications show spotty representation. For example, in the enterprise resource planning space, Baan NV is represented, but SAP AG and PeopleSoft are not. In the area of supply chain analytics, the only vendor represented is Manuguistics Inc.

A Definition of Data Warehousing

Bill Inmon
Bill Inmon is universally recognized as the "father of the data warehouse." He has over 26 years of database technology management experience and data warehouse design expertise, and has published 36 books and more than 350 articles in major computer journals. His books have been translated into nine languages. He is known globally for his seminars on developing data warehouses and has been a keynote speaker for every major computing association. Before founding Pine Cone Systems, Bill was a co-founder of Prism Solutions, Inc.

Ralph Kimball
Ralph Kimball was co-inventor of the Xerox Star workstation, the first commercial product to use mice, icons, and windows. He was vice president of applications at Metaphor Computer Systems, and founder and CEO of Red Brick Systems. He has a Ph.D. from Stanford in electrical engineering, specializing in man-machine systems. Ralph is a leading proponent of the dimensional approach to designing large data warehouses. He currently teaches data warehousing design skills to IT groups, and helps selected clients with specific data warehouse designs. Ralph is a columnist for Intelligent Enterprise magazine and has a relationship with Sagent Technology, Inc., a data warehouse tool vendor. His book "The Data Warehouse Toolkit" is widely recognized as the seminal work on the subject.



In order to clear up some of the confusion that is rampant in the market, here are some definitions:

Data Warehouse:

The term Data Warehouse was coined by Bill Inmon in 1990, which he defined in the following way: "A warehouse is a subject-oriented, integrated, time-variant and non-volatile collection of data in support of management's decision making process".

He defined the terms in the sentence as follows:

*

Subject Oriented: Data that gives information about a particular subject instead of about a company's ongoing operations.
*

Integrated: Data that is gathered into the data warehouse from a variety of sources and merged into a coherent whole.
*

Time-variant: All data in the data warehouse is identified with a particular time period.
*

Non-volatile: Data is stable in a data warehouse. More data is added but data is never removed. This enables management to gain a consistent picture of the business.

(Source: "What is a Data Warehouse?" W.H. Inmon, Prism, Volume 1, Number 1, 1995). This definition remains reasonably accurate almost ten years later. However, a single-subject data warehouse is typically referred to as a data mart, while data warehouses are generally enterprise in scope. Also, data warehouses can be volatile. Due to the large amount of storage required for a data warehouse, (multi-terabyte data warehouses are not uncommon), only a certain number of periods of history are kept in the warehouse. For instance, if three years of data are decided on and loaded into the warehouse, every month the oldest month will be "rolled off" the database, and the newest month added.

Ralph Kimball provided a much simpler definition of a data warehouse. As stated in his book, "The Data Warehouse Toolkit", on page 310, a data warehouse is "a copy of transaction data specifically structured for query and analysis". This definition provides less insight and depth than Mr. Inmon's, but is no less accurate.

Data Warehousing:

Components of Datawarehousing

Data warehousing is essentially what you need to do in order to create a data warehouse, and what you do with it. It is the process of creating, populating, and then querying a data warehouse and can involve a number of discrete technologies such as:

*

Source System Identification: In order to build the data warehouse, the appropriate data must be located. Typically, this will involve both the current OLTP (On-Line Transaction Processing) system where the "day-to-day" information about the business resides, and historical data for prior periods, which may be contained in some form of "legacy" system. Often these legacy systems are not relational databases, so much effort is required to extract the appropriate data.
*

Data Warehouse Design and Creation: This describes the process of designing the warehouse, with care taken to ensure that the design supports the types of queries the warehouse will be used for. This is an involved effort that requires both an understanding of the database schema to be created, and a great deal of interaction with the user community. The design is often an iterative process and it must be modified a number of times before the model can be stabilized. Great care must be taken at this stage, because once the model is populated with large amounts of data, some of which may be very difficult to recreate, the model can not easily be changed.
*

Data Acquisition: This is the process of moving company data from the source systems into the warehouse. It is often the most time-consuming and costly effort in the data warehousing project, and is performed with software products known as ETL (Extract/Transform/Load) tools. There are currently over 50 ETL tools on the market. The data acquisition phase can cost millions of dollars and take months or even years to complete. Data acquisition is then an ongoing, scheduled process, which is executed to keep the warehouse current to a pre-determined period in time, (i.e. the warehouse is refreshed monthly).
*

Changed Data Capture: The periodic update of the warehouse from the transactional system(s) is complicated by the difficulty of identifying which records in the source have changed since the last update. This effort is referred to as "changed data capture". Changed data capture is a field of endeavor in itself, and many products are on the market to address it. Some of the technologies that are used in this area are Replication servers, Publish/Subscribe, Triggers and Stored Procedures, and Database Log Analysis.
*

Data Cleansing: This is typically performed in conjunction with data acquisition (it can be part of the "T" in "ETL"). A data warehouse that contains incorrect data is not only useless, but also very dangerous. The whole idea behind a data warehouse is to enable decision-making. If a high level decision is made based on incorrect data in the warehouse, the company could suffer severe consequences, or even complete failure. Data cleansing is a complicated process that validates and, if necessary, corrects the data before it is inserted into the warehouse. For example, the company could have three "Customer Name" entries in its various source systems, one entered as "IBM", one as "I.B.M.", and one as "International Business Machines". Obviously, these are all the same customer. Someone in the organization must make a decision as to which is correct, and then the data cleansing tool will change the others to match the rule. This process is also referred to as "data scrubbing" or "data quality assurance". It can be an extremely complex process, especially if some of the warehouse inputs are from older mainframe file systems (commonly referred to as "flat files" or "sequential files").
*

Data Aggregation: This process is often performed during the "T" phase of ETL, if it is performed at all. Data warehouses can be designed to store data at the detail level (each individual transaction), at some aggregate level (summary data), or a combination of both. The advantage of summarized data is that typical queries against the warehouse run faster. The disadvantage is that information, which may be needed to answer a query, is lost during aggregation. The tradeoff must be carefully weighed, because the decision can not be undone without rebuilding and repopulating the warehouse. The safest decision is to build the warehouse with a high level of detail, but the cost in storage can be extreme.

Process Manufacturing: Industry Specific Requirements Part One: Introduction

Traditionally, manufacturing is categorized by two methods: process and discrete. Many differences exist, but most can be grouped into two areas: those derived from material issues and those derived from production issues.

Process materials are different than discrete materials. Process materials are powder, liquids or gases; they must be confined; and they are more difficult to accurately measure. Process materials are close to their natural sources (farms, mines, etc.) and therefore, are of inconsistent quality. Inconsistent quality means extensive quality procedures, segregation (lot control), restriction of use (this lot is okay for one customer but not another), and usually the inclusion quality attributes as part of their inventory definition. Process materials vary with time. They get better, they get worse, and they change their identity.

Production issues give us the simplest definition of process manufacturing. Specifically, once you produce your finished product, you cannot distill it back to its basic ingredients. Have you ever attempted to return orange juice back to its original water, sugar, sodium, and, of course, oranges or extract the pigments out of paint? Conversely, you can disassemble a car back to its tires, spark plugs, carburetor, and engine block. There are similar components in process and discrete manufacturing such as ingredients versus parts; formulas versus bill of materials; several units of measure (i.e., pounds, ounces, and liters) versus EA (each).

There are, however, subtle differences. Process manufacturing is scalable. For example, if the formula calls for a 1,000 pounds of oranges but you only have 500 pounds, you can still make orange juice; just not as much. If you only have three tires, you are going to have wait for the fourth tire before the car can start rolling off the production line. In process, you tend make product in bulk or batches as in a vat of coke or a 500-gallon tanks of solvent and then pack it off to fulfill customer orders. On the other hand, in discrete manufacturing you would expect to see one computer at a time coming down the production line.

For a quick refresher on process manufacturing, peruse the articles, Process Manufacturing: A Primer or What Makes Process Process.

The remainder of this article focuses on process manufacturing. However, to say process manufacturing functions are the same in all industries is tantamount to saying that a Ferrari and a Ford truck are simply means of getting from point A to point B. Just as you would not use a Ferrari to haul lumber, aspects of process manufacturing cannot be applied equally and with the same importance to all industries. This article looks at the unique requirements of process manufacturing in three industries: food and beverage, chemical, and a hybrid industry, textiles. One way or another, these requirements must be satisfied. If a software vendor can provide this satisfaction, your organization's anxiety level concerning the implementation of enterprise-wide systems can be significantly reduced.

If you are not in these industries, you can stop reading. No, wait! Perhaps, by understanding how a particular requirement or aspect of process manufacturing relates to one of these industries you may get a better understanding or insight on how it can be applied in your company. Whew! Thought that I had lost you! Glad you're back.

Editor's Note: For the purpose of this article, process and continuous-flow manufacturing are treated as synonymous. Continuous-flow manufacturing is the eradication of product stagnation in and between processes. Once a product has entered the manufacturing process, it moves on without having to be stored. Special considerations to establish a continuous-flow operation, such one-piece-at-a-time production and multi-process handling, , will not be addressed in this article.

This is Part One of a three-part note.

Part Two discusses process manufacturing requirements for the chemical industry.

Part Three discusses process manufacturing requirements for the textile industry and provides a summary.

Food and Beverage Industry

As you might expect, any industry that affects the health and welfare of the human race is bound to have special needs and requirements. With the incidents of Mad Cow Disease and the sudden and seemingly continuous preoccupation with the Atkins Diet, the most insignificant requirement for the food and beverage (F&B) industry is in the area of quality control (QC). While your customers may have their own special quality requirements, first and foremost, conformance must be established and verified with external agencies, such as (in the US) the Food and Drug Administration (FDA) and Bureau of Alcohol, Tobacco and Firearms (ATF) or your product will never reach the market. Consequently, integration with these external sources and frequent changes would be a critical element of the QC function. As you go further back into the supply chain process, the QC function must extend and usually starts with the supplier. Regardless, as the producer of a finished product, the responsibility for quality is joint and several which gives little allowance as to where the defect occurred in the supply chain. Look for software that seamlessly integrates with external agencies regulating your particular segment of the F&B industry.

Of special note is the US Bio-terrorism Act of 2002. This act places a series of new requirements on F&B companies. Most, including the authors, think that compliance with the Bio-terrorism Act is not possible without computerization of both the production process and the supply chain.

Once the regulated and external requirements are satisfied, there are customer and ingredients-related QC specifications that must be addressed. If your company is producing a finished product that is an ingredient into your customer's product, additional QC compliance is typically required. This could be for nutritional or ethnic considerations. Consequently, the setup of the QC function within the software must be flexible and adaptable.

The accurate statement of the QC specifications for the ingredients can also come into play. Going back to the orange juice example, the acidity of the oranges determines the amount of other ingredients (sugar, water, etc.) that may have to be adjusted to counteract the pH level. The pH level, recorded in the QC process, will therefore impact the product's specifications but, equally important, effect the "on the fly," one-time formula modification. Other QC-related requirements, that should be self-explanatory, include

• Nutritional reporting and labeling
• Taste QC specifications
• Color consistency QC specifications
• Shelf life longevity and reporting

Having worked in the food processing industry, the most terrifying words that you can hear on a Friday afternoon are, "This hamburger or soda tastes funny!" Your weekend, and possibly your livelihood, could be ruined and until you can dispel or confirm the damaging insinuation, an F&B organization is living in anticipatory paralysis. The fear stems from the negative financial impact on the company's image and customer base. Consequently, product recallability is an essential.

The Bio-terrorism Act of 2002 spells out detailed requirements which are often referred to as "one up and one down" tracking. This act also calls for the appropriate records within four hours from the receipt of a request from the FDA. Furthermore, recallability implies isolating and locating the defective product to an absolute minimum with dead-on certainty. To achieve this objective, "bullet proof" lot and sublot tracking is needed. This is easier said than done and can be an extremely time consuming process. However, certain attributes of lot/sublot tracking in the software can expedite the recording and tracking functions and help to eliminate damaging fallout.

First, there is lot to sublot inheritance. This means that characteristics of a lot are transferred automatically to the sublots contained within the lot. In so doing, the characteristics of bulk quantities of meat or oranges, for example, used to make hamburger patties or juice, respectively, are retained or inherited by the boxes and crates of the finished product. As a result, the recording of sublots places less hardship on the production line personnel and is less prone to recording mistakes or errors of omission.

Secondly, lot tracking should follow the product through any re-work processes. Even with undergoing a re-working process, the original lot and sublot characteristics should not be lost unless the re-work makes these characteristics meaningless.

Finally, lot and sublot tracking must be able to remain intact until the product arrives at the customer's location. This is the only way a complete recall can be accomplished and the questionable product returned to the manufacturers. Software gaps, preventing any one of these three requirements from being satisfied, brings the entire recall process into question and would require significant custom coding or administrative procedures to be filled.

Other Operational Issues

There are several additional operational issues that any self-respecting F&B software should be able to address. In addition to accommodating picking strategies such as LIFO (last in first out), FIFO (first in, first out), and FEFO (first expire, first out) the software must account for the perishability of the ingredients as well as the finished product. Consequently, taking into account the expiration date is key when determining picking priorities.. Some customers also demand strict rotation where the supplier can never ship product that is older than the last shipment.

For some manufacturers, private labels represent a significant segment of a F&B production run. Using the private label concept, large supermarkets utilize the value of name recognition to provide products under their own label like Safeway, Albertson's, Royal Ahold, and Tesco. Because of the large quantities required by these customers, manufacturers usually cannot wait until the order is on hand to start up the production line. Alternately, if the raw ingredient is only available in season (vegetables in August for example), the entire year's demand must be processed in a limited time period. Accordingly, a food processor will create unlabelled products. Labeling will only be completed after the sales order is received and confirmed.

Because of their extended shelf life, cooked, canned goods lend themselves well to this type of production. Sealed aluminum cans remain on an inventory shelf for up to twelve months while waiting for labeling. Hence, the terms, "brite stock" or "shiny stock" were created to refer to this type of stock. To be able to accommodate requirements lot and sublot tracking must extend and be maintained within the brite stock. Also, the manufacturing process must be able to be separated into two stand alone, independent processing runs. One would be for the production run to make the brite stock and a second, a packaging run to label and ship the product.

Catch weight or random weight is a common, and non-negotiable requirement for some food categories, particularly with meats. While meat and poultry products may be advertised for $50 a box, tin, or drum, the invoiced price is based on the actual, not estimated or expected weight of the product. Accordingly, not only does the software have to track the total weight, including packaging weight, to calculate shipping charges, it must also track the catch weight for pricing. While the concept may be simple to comprehend, its application may not that easy. However, this is an industry practice that cannot be ignored in some categories; it is the way business is done.

Some food companies are in the "disassemble" business. These companies grow or acquire one raw material and make many products from this single raw material. For example, a chicken processor may buy live chickens to make many different parts. An apple processor buys many different grades and sizes of apples, sorts them, and processes them into many different products. In contrast, discrete companies buy many different parts to make one end item and the bill of material was designed for this purpose. In process manufacturing, when one raw material is made into many end-items, a formula or recipe (process's equivalent to the discrete bill of material) is being asked to do something for which it was not designed. Consequently, a formula must have the flexibility and tensile strength to be changed rapidly and still conform the existing resources and routings on the plant floor. For example, it may be a "game time" decision on how to process a batch of apples to maximize the product yield. The software must be able to accommodate these changes through formula and routing modifications and still stay within the constraints of the plant floor. Of course, we want to maintain the integrity of the original formula and routing.

Additionally, companies making multiple end items from a single incoming ingredient have a series of requirements that must be addressed. In place of a traditional bill of materials, they require a model that accounts for multiple outputs, often called by-products, co-products, and waste. With this model, companies require scheduling which reflects the logic of the plant. Are they scheduling a quantity of end item, end items, units of inputs or hours of processing time? All are common. If planning is to be used, will the planning system deal with independent demand for multiple items or limit demand to a single item? Costing functionality must reflect the various methods used today. Some end items are priced at market (the currently available price for that item), produced waste items should have their cost of disposal charged back to the process (and therefore products) that produced the waste. Finally, multiple end items may require that the cost of the entire process to be split based upon a percentage split of cost. Since producing multiple end items means that we may have the same item as both a consumed item and a produced item, the system must deal with recycles. Recycles can have a significant impact scheduling, planning, and costing.

Other aspects of the F&B industry that you must be aware of are:

• Flexible packaging alternatives (i.e. consider the different ways you can purchase soda).
  
  
• Re-pack functionality (i.e. don't have soda in 1 liter bottles but can re-pack 55-gallon drums).
  
  
• Bulk storage using tanks and silos and the need to maintain, record, and track temperature and spoilage attributes.
  
  
• Special needs of fresh, chilled, and frozen ingredients and products.
  
  
• Container management for beverages.

This concludes Part One of a three-part note.