Business Forecasts with Artificial Intelligence

Today's business world is driven by customer demand. Unfortunately, the patterns of demand vary considerably from period to period. This is why it can be so challenging to develop accurate forecasts. Forecasting is the process of estimating future events, and it is fundamental to all aspects of management. The goals of forecasting are to reduce uncertainty and to provide benchmarks for monitoring actual performance. Emerging information technologies and artificial intelligence (AI) techniques are being used to improve the accuracy of forecasts and thus making a positive contribution to enhancing the bottom line.

A new generation of artificial intelligence technologies have emerged that hold considerable promise in helping improve the forecasting process including such applications as product demand, employee turnover, cash flow, distribution requirements, manpower forecasting, and inventory. These AI based systems are designed to bridge the gap between the two traditional forecasting approaches: managerial and quantitative.

Forecasting Approaches

Generally speaking, forecasts are based on quantitative analysis, qualitative analysis or a combination of both. Often quantitative forecasting is referred to as objective analysis while qualitative forecasting is called managerial or judgmental analysis. Typically, there is tension between these two approaches. Quantitative forecasts, which are often favored by operations, tend to be developed using a bottom up approach while managerial-based forecasts, usually preferred by the marketing group, are approached from a top down perspective. For example, a primary marketing goal is to insure adequate supply while operation's focus is on minimizing inventory. The resolution of these two approaches is how forecasting errors occur and presents an opportunity for using artificial intelligence methods. Quantitative forecasting can be characterized by one of the two basic techniques:

• Time Series - The future will tend to look and behave like the past. For example, gasoline prices for the next six months will continue along the same lines as they have over the past six months.
• Relational - The future is dependent on the direction of a variety of factors. For example, new housing starts might be a function of interest rates and local weather conditions.

A time series is a set of data points recorded over successive time periods. Examples include monthly billables, weekly unit product demand and quarterly inventory levels and stock prices. A relational database consists of the recording of several variables for a number of observations. For example, a financial relational database could consist of revenues, earnings and assets for the Fortune 500.

The following graphic highlights the typical forecasting process. The resultant forecasts are evaluated by comparing predictions with actual results. This assessment is accomplished by examining the error terms. An error term is the difference between the prediction and the actual outcome. Based on an error assessment, the forecasting process is continually updated through the adjustment of model inputs.

Forecasting Approaches

Generally speaking, forecasts are based on quantitative analysis, qualitative analysis or a combination of both. Often quantitative forecasting is referred to as objective analysis while qualitative forecasting is called managerial or judgmental analysis. Typically, there is tension between these two approaches. Quantitative forecasts, which are often favored by operations, tend to be developed using a bottom up approach while managerial-based forecasts, usually preferred by the marketing group, are approached from a top down perspective. For example, a primary marketing goal is to insure adequate supply while operation's focus is on minimizing inventory. The resolution of these two approaches is how forecasting errors occur and presents an opportunity for using artificial intelligence methods. Quantitative forecasting can be characterized by one of the two basic techniques:

• Time Series - The future will tend to look and behave like the past. For example, gasoline prices for the next six months will continue along the same lines as they have over the past six months.
• Relational - The future is dependent on the direction of a variety of factors. For example, new housing starts might be a function of interest rates and local weather conditions.
• The following process outlines a plan for improving forecast accuracy using artificial intelligence support systems:

  1. Evaluate and characterize the current forecasting system.

  2. Measure the current level of error.

  3. Compare error levels with industry norms.

  4. Specify new requirements.

  5. Characterize the economic impact of improved forecasts.

  6. Identify alternative AI forecasting options.

  7. Select best approach(s).

  8. Develop implementation schedule.

  9. Identify potential bottlenecks and problem areas.

  10. Implement new system and monitor performance.

Future of Business intelligence in Retail:

The Future of BI in Retail:

BI will be defined by the retailers that have figured out how to maximize customer satisfaction and profitability with the right combination of quality products, friendly and efficient service, unique value, a differentiated shopping experience, and a business model that truly serves its community -- locally and globally. How will this be accomplished? It starts with understanding the customer and then linking that insight into every decision that is made, from merchandising to marketing to distribution to store operations to finance, so that retailers can predict how to best serve their customers' ever-changing needs and desires.

Our vision for the future of retail BI provides for that very scenario, through our intelligence platform and our solutions for customer, merchandise, operations, and performance intelligence that are combined in a suite designed to equip retailers to become truly innovative.

A solution seeking to use customer behavioral data to make better merchandising or marketing decisions needs to interface with sales transaction systems, loyalty systems, in-house credit systems, coupon redemption systems, catalog and Internet customer data systems, and so forth. A system that recommends optimized price changes should interface with the price management system, the item master, the system that generates labels, etc.

t is important to note that a good BI solution will be able to integrate with any other system or platform. That said different BI solutions need to interface with different operational systems for different purposes.

A solution seeking to use customer behavioral data to make better merchandising or marketing decisions needs to interface with sales transaction systems, loyalty systems, in-house credit systems, coupon redemption systems, catalog and Internet customer data systems, and so forth. A system that recommends optimized price changes should interface with the price management system, the item master, the system that generates labels, etc.

There must be a closed-loop interface between the operational systems that retailers rely upon to conduct day-to-day business and the BI systems that help them conduct that business more efficiently and profitably.

Marketing -- By understanding customers better -- whether by profiling, segmenting, gauging propensity to respond, or using market basket analysis -- retailers can create better-defined targeted campaigns, reducing expenses (printing, paper, postage) while increasing response rates, revenues, and gross margins. Also, as retailers gain a better understanding of their customers' buying behavior, this analysis can then be used to create more effective merchandising plans for the next season.

Operations -- Understanding and predicting changes in demand -- by hour, by day, by location, by promotion, by price change -- mean that the store floors, the catalog call centers, and the fleet crews delivering replenishment orders from the DC to the store are all appropriately staffed. This understanding also leads to optimal productivity since store-level human capital costs can be scheduled better and managed more efficiently.

Specific Areas in Business intelligence solutions

Operational and transactional systems such as merchandise management, ERP (enterprise resource planning), and POS, are very good at what they do -- organizing huge amounts of operational data and transactions. These systems can tell retailers what has happened in their business and what their customers have done -- last week, last month, and last year.

It's critical, however, for retailers to understand what will happen: what the demand will be for a select assortment of merchandise, what impact an incremental price change will have on demand, which floor plan will sell more designer shoes, which customers will respond to a direct mail or catalog offer.

Real value comes from systems that go beyond the limitations of operational software alone, systems that can take operational data and create enterprise intelligence and predictive insights.

These BI systems must combine data management (consolidating, organizing, and cleansing huge amounts of disparate data from varying systems and platforms) with predictive analytics (data mining, forecasting, optimization). When they do, retailers can make sense of customer, product, supplier, and operational data and draw insights that will help them run their businesses better and more profitably.

Leading retailers around the globe -- like Wal-Mart, Foot Locker, Staples, Williams-Sonoma, and Amazon.com and many others -- have begun using BI and analytics to make an array of strategic decisions. These include where to place retail outlets, how many of each size or color of an item to put in each store, and when and how much to discount. The effects of these decisions can save or generate millions of dollars for retailers.

The Strength of the Market for BI in Retail Today:

The market is very strong and getting stronger. While it is difficult to find a comprehensive suite of retail-specific BI offerings that spans the spectrum from competitive intelligence to merchandise planning and optimization (product, price, promotion, and placement) based on customer insight, to knowing how to maximize the ROI on the next marketing campaign, to understanding where to build the next store, to reducing supply chain costs. Retailers are telling us over and over that they are seeking a single, stable, reliable, and proven provider of superior BI solutions. They are implementing projects that span multiple years and will deliver value for years to come.

The Retailers that are Realizing the Most Benefits from BI:

We find that the retailers that are realizing the most significant returns on their investments are those that take a purposeful, pragmatic approach to establishing an intelligence platform upon which to base all other BI solutions. A single, reliable demand forecast, for instance, can also be used in merchandising, marketing, logistics, store operations, call center staffing, etc., for operational benefit. BI that remains segmented by functional area can provide some value, but retailers can realize a much larger return by building the foundation upon which the rest of the house will stand. This is true of both top-tier and midmarket retailers, regardless of segment.

Specific Areas in Which Retailers can Benefit Most Include:

Merchandising -- This is clearly the most important area of a retailer's business and an area where retailers are beginning to exploit the full value of BI. Analysis of past performance, combined with plans and forecasts of future customer behavior, leads to more accurate initial allocations of merchandise across channels and stores. Assortment and size optimization that are based on customer demand patterns ensure that the correct assortments, size, and case-pack distributions get sent to the correct stores. Daily price, promotion, and markdown optimization ensures that items are priced for optimal profitability, both preseason and in season. Space automation and optimization ensure that departmental sales and profit per square foot are maximized, and products are given the correct inventory and space on the shelf or on the rack. Optimized fulfillment ensures that products are allocated or replenished based on demand. Accurate analysis also results in a more efficient use of manpower in picking, packing, and shipping the first wave of product, while minimizing additional, costly payroll expenses to facilitate transfers between stores, vendor returns, changing signage and labels for markdowns, and otherwise correcting mistakes.

Business intelligence solutions for the retail industry

Traditionally, the retail industry has lagged behind other industries in adopting new technologies, and this holds true in its acceptance of BI technology. Some industries, such as financial services, have become very sophisticated in using BI software for financial reporting and consolidation, customer intelligence, regulatory compliance, and risk management. However, retailers are quickly catching up and beginning to recognize the many areas of BI that can be applied specifically to their businesses.

The Strength of the Market for BI in Retail Today:

The market is very strong and getting stronger. While it is difficult to find a comprehensive suite of retail-specific BI offerings that spans the spectrum from competitive intelligence to merchandise planning and optimization (product, price, promotion, and placement) based on customer insight, to knowing how to maximize the ROI on the next marketing campaign, to understanding where to build the next store, to reducing supply chain costs. Retailers are telling us over and over that they are seeking a single, stable, reliable, and proven provider of superior BI solutions. They are implementing projects that span multiple years and will deliver value for years to come.

The competitive game is changing for retail. As the industry continues to consolidate, retailers have begun to realize that using technology to better understand customer buying behavior, to drive sales and profitability, and to reduce operational costs is a necessity for long-term survival.

Retailers are now paying significant attention to BI software, specifically in the areas of merchandise intelligence (including merchandise planning, assortment, size, space, price, promotion, and markdown optimization), customer intelligence (including marketing automation, marketing optimization, and market basket analysis), operational intelligence (including IT portfolio management, labor optimization, and real estate site selection), and competitive intelligence. There are many factors that have led retailers to adopt BI software: increased competition, the need to squeeze more profitability out of less space, prevalent credit card usage, the Internet's role as an alternative sales channel, the popularity of loyalty cards, and soon, RFID (radio frequency identification). These milestones have created a wealth of data that retailers are now beginning to appreciate and use.

Within individual companies, we view the history of BI in retail through a method that we devised to describe the status of any company's evolution toward becoming an intelligent enterprise. We believe that organizations pass through five fundamental stages as they advance in their use of BI as a competitive differentiator:

Operate -- At this most basic level are the companies rife with information mavericks: the guys in basement offices hammering away on desktop spreadsheets. If they go, the knowledge goes with them. There are no processes, and each request becomes an ad hoc data rebuild, resulting in multiple versions of the truth, with the likelihood of a different answer to any one question every time it is asked.
Consolidate -- At this stage, a company has pulled together its data at the departmental level. Here, a question gets the same answer every time, at least within the department. However, departmental interests and interdepartmental competition can skew the integrity of the output and result in multiple versions of the truth.

Integrate -- At this point in the evolution, a company has adopted enterprise-wide data and bases its decisions on this more complete information. This company is beginning to have a true awareness of additional opportunities for the use of BI to improve processes and profits.

Optimize -- At this stage, the company's knowledge workers are very focused on incremental process improvements and refining the value-creation process. Everyone understands and uses analysis, trending, pattern analysis, and predictive results to increase efficiency and effectiveness. The extended value chain becomes increasingly critical to the organization, including the customers, suppliers, and partners who constitute intercompany communities.

Innovate -- This level represents a major, quantum break with the past. It exploits the understanding of the value-creation process acquired in the optimize stage and replicates that efficiency with new products in new markets. Companies operating at this level understand what they do well and apply this expertise to new areas of opportunity, thus multiplying the number of revenue streams flowing into the enterprise. Armed with information and business process knowledge, organizations approaching the innovate level will introduce truly innovative products and services that reflect their unique understanding of the market, their internal strengths and weaknesses, and an unfailing flow of ideas from continuously engaged employees.
We are finding that most large retailers have reached or are approaching the integrate stage, with many making great strides toward the optimize and innovate levels. There is an enormous opportunity for the evolution to continue -- within every retail organization.

The Presence of BI in the Retail IT Infrastructure:

In the typical retail IT infrastructure, there are two fundamental categories of systems: transactional/operational systems, such as POS and purchase order management systems; and analytic/BI systems.

business inteligence innew era

As 21st century is on ramp we need to think about business processes to come up in 2012 onwards. It will be people oriented business processes .I mean the People whom company is holding. product will have little less importantance as every body will have product.But companies will diffrenciate with Human cepital behind it.