- Augmented Analytics
- Big Data Analytics
- Business Analytics
- Data Governance & Data Fabric Architecture
- Data-Driven Strategy & Transformation
- Performance Management & Improvement
- About us
- Insights Room
- Webroom & Events
Demand planning and forecasting are not stand-alone processes, but they must be integrated into other aspects of operations to provide value.
Demand planning and forecasting are not stand-alone processes, but they must be integrated into other aspects of operations in order to provide value. One of these processes is S&OP.
The authors J. Jayaraman, D. Natarajan, M. Romeri, & T. Zych, in their “Insights into supply chain operations strategy” define S&OP as “a process focused on ensuring a continuous alignment between demand, inventory, supply and manufacturing plans on one hand, and the relations between these tactical plans and the business plan on the other hand, in order to maximize operational performance”.
An Aberdeen Group study noted that more than 60% of Best-in-Class companies see the S&OP process as a strategic priority within their organization. This approach can be critical to a company’s success as it provides a useful decision-making tool for managing sales and operations. Below an insight on the main aspects that every organization should consider in order to take advantage in the best way of S&OP.
S&OP 4.0 has the ability to create a replicable process model that can match demand, supply, and finance to business goals. With this logic, the Demand project appears to be one of the three fundamental tools, in addition to industrial finance planning and operational planning.
Demand planning is the process that combines statistical forecasts and all internal and external market and customer information (both quantitative and qualitative), where human intervention plays a significant role, especially with regard to the integration of the information reported by the various functions during the meeting. (Demand planning. Risultati di un’indagine sul campo, Fabrizio Dallari e Alessandro Creazza Centro di Ricerca sulla Logistica – Università Cattaneo LIUC)
This tool makes the decider less subject to unexpected events as it requires a more scientific approach to the data.
The study conducted by the Cattaneo University LIUC highlights that over 60% of companies use a software application to make predictions, 23% don’t use any software, while the rest are adopting general purpose programs as Excel. Although companies recognize distributors and retailers as a key lever to improve forecasts, more than half of them are not committed to this. The average level of upstream and downstream collaboration is, in fact, very low for all surveyed companies (about 2 out of 10).
The main criticalities identified by the study are:
- a general misunderstanding about the meaning of statistical-predictive activity;
- an inaccurate statistical approach with unprocessed and reliable data.
In addition, the use of information technology systems is not synonymous with the actual use or even integration of functions. In fact, there are frequent cases in which different software used by different and off-the-shelf workings exist in the company, favoring duplication of processes and waste of resources, but also cases of failed implementations due to the lack of involvement of actual users, or cases of software under-utilization due to the lack of adequate statistical and computer skills. All the issues and criticalities that should spur companies in identifying a best practice to follow, improving the forward process. Demand Planning in S&OP 4.0 logic provides a tool of operations which aim is solving these problems and achieving a satisfactory result.
Model settings and on-demand analysis
This aspect of Demand Planning refers to the definition of a reference time range, the predisposition, and optimization of dependent and independent variables and regressors. It is also necessary to define the scope of analysis as a number of sales points or customers and its variation to refine future forecasting and planning.
Data preparation is a preliminary phase that aims to clear the raw data, exclude from trend analysis irregular observations (as seasonality, cyclicity or outliers), separate promo period in order to analyze the baseline forecast. Following this preparation, a preliminary statistical analysis is planned: time plot, position index, pattern.
This phase, the core of Demand Planning, plans to prepare a Forecast Model Battle, the processing of quantitative testing such as goodness-of-fit test and forecast accuracy. Once initial processing is made we commit ourselves to a top down or bottom up reconciliation of the data depending on the type of analysis being followed. Finally, specific analyzes for abnormal series, scheduled orders, new products are also provided.
Once the model has given results, they must be ordered, analyzed and interpreted in the right way. The model output aims at reaching out an outline expressed in reporting variables and specific KPIs to continuous monitoring of the parameters found, regressions on the future, estimation of the minimum levels to be guaranteed, and finally estimate the forecast error. Besides, it’s important to consider the need to handle cannibalized products equivalent to that of the company itself (for example, on promotions and launch of new products).
Therefore, the highest reliability of historical data must be sought first, so that baseline extrapolations can be equally reliable, to seek from the start a software-man integration, in the implementation and development phases in order to create confidence in the potential of the application.
INDUSTRIAL FINANCIAL PLANNING
Why is important for an industry to develop an Industrial Financial Project?
The aim of an IFP project is to build an all-in-one system, useful to process both Industrial Plan and Material Requirement Plan. Thanks to the output of these two plans it is possible to obtain an analysis of the Working capital and process as forecast on the Industrial Cash flow. This instrument allows a company to cut deadlines through his multi-user interface.
The Industrial Plan through an allocation algorithm can decide when the production of an item should start based on the available capacity, coherently with the lead time of the different production steps. Before this allocation, it is possible to set a weekly capacity calendar for the plants or the supplier.
Material procurement plan
The Material Procurement Plan is based on the BOM explosion and on the allocation of the Raw Material Request on the suppliers. The plan should be done on clusters of raw materials, avoiding excessive granularity.
Working capital and cash flow
When these two plans are complete and the Receivables are known, the system is able to compute the number of payables, the inventory variance for every time bucket and finally the Industrial Cash Flow.
When we talk about “supply chain” we mean a wider concept than “logistics”; in fact, we can define logistics as a planning orientation and framework that aims to create a single plan for the flow of products, while supply chain can be defined as the building of a framework that aims to coordinate the processes between the organization and other entities, like suppliers or customers.
How to bring the right product to the right place, at the right time, price and cost?
The first competitive advantage comes from the optimization of network management. First of all, each organization should evaluate the real need for each transit point, and analyze the efficiency in terms of delivery time of products. Besides, it’s important to evaluate the cost of transportation derived from each transit point to the customer, in order to monitor its trend during the year.
With this information, the organization could decide to merge inefficient transit point allocated in small areas or find more efficient locations than the ones already used.
Inventory management means the management of inventory and stock. As an element of supply chain management, it includes aspects as controlling and overseeing ordering inventory, storage of inventory, and controlling the amount of product for sale. There are many different models of stock management, and usually, each organization adopts one model for its entire set of products; choosing the model that fits best for a subgroup of products could improve the competitive advantage.
The subgroups should be created considering aspects like consuming deadline, storage temperature or perishability. The implementation of lean inventory management understood as a systematic approach to enhancing value in a company’s inventory by identifying and eliminating waste of materials, effort and time through continuous improvement in pursuit of perfection is suggested mostly for perishable products.
Effective logistics management is incomplete without proper warehouse management, which must consider the different characteristics of goods in stock to better manage movements throughout the logistics network.
Significant advantages can be achieved in terms of productivity, speed, and accuracy with the benefit of a good warehouse management system (WMS). The WMS normally interfaces with the company’s main transaction system to access information and will feed back information such as goods received and dispatched. The WMS is used to control all the operations in the warehouse and issues instructions to subsidiary stock keeping sites.
In any inventory system which includes more than one type of item, an internal classification must be done in order to correctly evaluate the importance of the stocked goods (overall value, higher usage rate, expiring dates...). Typically Warehouse management develops a ranking system often based on inventory usage value and typically a small part of the overall inventory will be the most used according to the 80/20 rule which is often used to create a macro cluster of goods and organize monitoring and controlling processes (ABC Ranking system).
A way of measuring inventory rate of usage in comparison to the inventory availability is the stock coverage ratio, which measures the duration of time before depletion of stock levels according to a usage parameter. Reciprocally, a stock turn parameter can be used which analyses how often the stock is used up (‘turns’) in a period. These ratios are often declined throughout the supply chain to match the below requirement of the network and represents the desired target object.
The last part of warehouse management consists in analyzing target coverage ratios and managing stock movement not only in a vertical way but also horizontally, evaluating cost-opportunity between losing sales and accounting for transportation and logistical costs of moving goods between distribution channels to amend over or understocking situations.
Dark data for lost sales and analysis
All the considerations of the previous points are connected to the analysis of data usually done in each organization. But what about the pieces of information hidden behind dark data? Gartner defines dark data as the information assets that organizations collect, process and store during regular business activities, but generally fail to use for other purposes. Dark data is still unknown from most of the organizations but it represents the main source of a competitive advantage for the future of S&OP 4.0.
This article has been written by Nicola De Beni, Fabio Giordano, Giulia Groff, Michele Martignoni, Francesco Rigatti, Silvia Tamellini. All of them are SDG Group consultants.
• Le tante sfumature del Sales & Operations Planning, Logistica Efficiente - Il portale della Logistica
• Demand planning Risultati di un’indagine sul campo, Fabrizio Dallari e Alessandro Creazza Centro di Ricerca sulla Logistica – Università Cattaneo LIUC, 2006
• Pentz, K., The Difference between Demand Planning, Forecasting and S&OP, Institute of Business Forecasting & Planning
• Rushton, A., Croucher, P., Baker, P. 2010. The Handbook of Logistics & Distribution Management. 4th ed. London: Kogan Page Limited.
• Slack, N., Chambers, S., Johnston, R. 2010. Operations Management. 6th ed. Edinburgh: Pearson Education Limited.