This paper puts forward an approach on how to create duplicate records from one Adverse Event (AE) datum based on study treatments. In order to fulfill this task, one flag was created to check if we need to produce duplicate records depending on an existing AE in different treatment periods. If yes, then we create these duplicate records and derive AE dates for these duplicate records based on treatment periods or discontinued dates.

Predicting the profitability of future sales orders in a price-sensitive, highly competitive make-to-order market can create a competitive advantage for an organization. Order size and specifications vary from order to order and customer to customer, and might or might not be repeated. While it is the intent of the sales groups to take orders for a profit, because of the volatility of steel prices and the competitive nature of the markets, gross margins can range dramatically from one order to the next and in some cases can be negative. Understanding the key factors affecting the gross margin percent and their impact can help the organization to reduce the risk of non-profitable orders and at the same time improve their decision-making ability on market planning and forecasting. The objective of this paper is to identify the best model amongst multiple predictive models inside SAS^® Enterprise Miner™, which could accurately predict the gross margin percent for future orders. The data used for the project consisted of over 30,000 transactional records and 33 input variables. The sales records have been collected from multiple manufacturing plants of the steel manufacturing company. Variables such as order quantity, customer location, sales group, and others were used to build predictive models. The target variable gross margin percent is the net profit on the sales, considering all the factors such as labor cost, cost of raw materials, and so on. The model comparison node of SAS Enterprise Miner was used to determine the best among different variations of regression models, decision trees, and neural networks, as well as ensemble models. Average squared error was used as the fit statistic to evaluate each model's performance. Based on the preliminary model analysis, the ensemble model outperforms other models with the least average square error.

Six Sigma is a business management strategy that seeks to improve the quality of process outputs by identifying and removing the causes of defects (errors) and minimizing variability in manufacturing and business processes. Each Six Sigma project carried out within an organization follows a defined sequence of steps and has quantified financial targets. All Six Sigma project methodologies include an extensive analysis phase in which SAS^® software can be applied. JMP^® software is widely used for Six Sigma projects. However, this paper demonstrates how Base SAS^® (and a bit of SAS/GRAPH^® and SAS/STAT^® software) can be used to address a wide variety of Six Sigma analysis tasks. The reader is assumed to have a basic knowledge of Six Sigma methodology. Therefore, the focus of the paper is the use of SAS code to produce outputs for analysis.

The two primary objectives of multi-tiered causal analysis (MTCA) are to support and evaluate business strategies based on the effectiveness of marketing actions in both a competitive and holistic environment. By tying the performance of a brand, product, or SKU at retail to internal replenishment shipments at a point in time, the outcome of making a change to the marketing mix (demand) can be simulated and evaluated to determine the full impact on supply (shipments). The key benefit of MTCA is that it captures the entire supply chain by focusing on marketing strategies to shape future demand and to link them, using a holistic framework, to shipments (supply). These relationships are what truly define the marketplace and all marketing elements within the supply chain.

In this era of bigdata, the use of text analytics to discover insights is rapidly gainingpopularity in businesses. On average, more than 80 percent of the data inenterprises may be unstructured. Text analytics can help discover key insightsand extract useful topics and terms from the unstructured data. The objectiveof this paper is to build a model using textual data that predicts the factorsthat contribute to downtime of a truck. This research analyzes the data of over200,000 repair tickets of a leading truck manufacturing company. After theterms were grouped into fifteen key topics using text topic node of SAS^® TextMiner, a regression model was built using these topics to predict truckdowntime, the target variable. Data was split into training and validation fordeveloping the predictive models. Knowledge of the factors contributing todowntime and their associations helped the organization to streamline theirrepair process and improve customer satisfaction.