SourceForge recently caught up with Greg V. Hartunian, CEO of Smart Software, to talk about today’s forecasting and planning solutions and discuss the business benefits of inventory optimization. Greg also offers some advice to help enterprises reduce standing inventory, increase parts availability, and reduce the need for and associated costs of emergency transshipment to close gaps in their supply chain
Q. Tell me about Smart Software.
GVH: Smart Software enables organizations to plan inventory policies and demand forecasts quickly, easily, and accurately. Using data, automation and analytics, we ensure stocking levels that will maximize service and minimize costs. We are headquartered in Belmont, Massachusetts, USA just outside of Boston.
Smart Software was founded in 1984 by Charles Smart, Thomas Willemain and Nelson Hartunian. They were true visionaries in the forecasting space and introduced the first-ever automatic statistical forecasting application for the PC. Our desktop software, SmartForecasts, has helped thousands of users forecast more accurately and yielded tens of millions in savings for our customers. It is still used by many customers and fully supported.
In 2015, we realized that the best way to help our customers was to develop a scalable web-based planning platform that would leverage our field-proven capabilities, make collaboration and reporting easier, and serve as a launching point for a host of new applications and innovations. Smart Inventory Planning & Optimization (IP&O) was launched in 2016. It is multi-tenant web platform hosted on AWS with 3 core apps: Demand Planner, Inventory Optimization, and Operational Analytics.
Q: What problems does your company and software solve?
GVH: We help our customers identify and prevent excess inventory, stock outs, and operational inefficiencies that result from inadequate planning. If a company has too much inventory and would like less of it, is always scrambling to fill orders or losing business due to frequent stockouts, or has a planning team that spends most of its time troubleshooting instead of planning, we can help. Smart finds the optimal balance between two competing extremes – the need to maintain high service levels to customers but keep inventory investments minimized. Finding this balance ensures that limited resources can be allocated properly across thousands of items and many locations, maximizing the organization’s return on inventory investment.
Q: Why is this important?
GVH: Inventory is very expensive. Typically, carrying costs associated with storing, ordering, and managing inventory represent 20-30% of the item’s unit cost annually. For every $1M in inventory it costs roughly $200,000- $300,000 annually to maintain. Conversely, an inventory that is too lean will result in frequent stockouts, lost sales, and eventually lost customers. I don’t have to tell you what happens to a business that can’t consistently fill orders in full or on time. Keeping inventory investments in check while maintaining high customer service levels is really hard to do and a constant balancing act. Without proper controls, the right people, and the right planning platform, problems will abound: excess inventory grows throughout the supply chain on some items, locking up vital working capital resulting in stock outs and lost sales on other items. This constrains the company’s growth and can cripple the business in hard times.
Q. What does your software do exactly?
GVH: In summary, the software develops statistically generated demand forecasts and optimal stocking parameters such as reorder points and reorder quantities and returns these values to your ERP or EAM systems to drive ordering, replenishment, and production planning. We do this in 6 steps:
1. First, Smart regularly imports up-to-date historical inventory transactions from your ERP or EAM system. The data processing classifies items based on volume/frequency, applies item replacements, accounts for preferred stocking location rules, and more.
2. Smart then uses Machine Learning and Probabilistic Forecasting models to simulate future demand and lead times. It determines the probability for each possible demand and lead time that might occur.
3. Next, Smart automatically identifies the optimal reordering parameters that perform the best after being stress-tested over thousands of simulated demands and lead times. Smart’s optimization logic prescribes the parameters and service levels that yield the best return for the business.
4. Predicted metrics such as service level, fill rate, holding costs, ordering costs, and stock out costs are then compared across multiple proposed and current policies helping identify items at risk of excess or shortage.
5. The collaborative workbench enables demand forecasts and stocking policies to be overridden, reviewed, approved, and finalized before being saved back to the ERP or EAM system. Forecasts can be analyzed at multiple levels of the forecast hierarchy.
6. Performance reports reveal forecast accuracy, historical and current service levels, stock out events, customer ship times, supplier lead times, on hand inventory, turns, and more. Users can track performance over time for any item, family, group, location, and more.
Q: How do companies typically grapple with this problem?
GVH: Not very well. Many organizations rely on manual processes that are overly reliant on rule of thumb math, outdated statistical methods, and “gut feel.” These approaches yield suboptimal policies that cause inventory costs to balloon and service performance to suffer. For example, it’s quite common for companies to base reorder points on a multiple of average demand. This doesn’t account for demand variability and will nearly always overstock predictable items and understock less predictable items.
Compounding the problem is the sheer volume of data – thousands of items stocked at multiple locations means planners don’t have the bandwidth to proactively review items on a regular basis. This results in outdated stocking parameters that leak money and further contribute to the problem.
We also observed that quite often, the organization’s planning knowledge including details on how a company sets policies and places orders are often not widely understood by the organization. This knowledge is often isolated to a small group of planners (or even a single planner) with a process that is baked into a complicated spreadsheet with little to no supporting documentation. This means the process isn’t portable, isn’t scalable, and places significant dependencies on a small number of people.
Q: How is this relevant in today’s climate where everything is so volatile?
GVH: It’s more important to plan properly today than ever. In 2020 companies are faced with SKU proliferation, more complex supply chains, increased demand/supply variability, and more demanding and less forgiving customers. Without the right people and planning foundation, most companies will struggle to keep up, putting their businesses at risk.
The current Covid-19 threat has highlighted the critical need for adaptable planning processes. The abilities to dynamically identify discontinuities in demand and supplier lead times, modify analytic approach and planning frequency, and incorporate business knowledge are especially critical and central to our solution.
Q: What industries do you serve and who are your current customers?
GVH: We have customers in nearly every industry since forecasting and inventory planning is a vertical application that can help any inventory carrying organization. If you have inventory, would like less of it and/or need better customer service, we can help. Our ideal customer has thousands of items, many of which have unpredictable, intermittent demand that isn’t accurately forecasted using traditional approaches. This means we tend to draw lots of interest from discrete manufacturers and distributors. We also have many customers in the MRO space that manage large spare parts inventory where it is mission-critical to have the right parts on hand to ensure machine/equipment uptime.
Distributors must be prepared fill orders right away or lose sales and customers. Manufacturers must be ready with an adequate supply of components to ensure on time production and shipment or face expensive expedites, emergency jobs, and lost sales. MRO organizations must ensure nearly 100% machine uptime or incur significant financial penalties from violating service level agreements.
Our better-known customers include FedEx TechConnect, WeatherTech, Hitachi High Technologies, Otis Elevator, Metro-North Railroad, and Disneyland Resorts.
Q: What is the most important differentiator Smart Software offers?
GVH: Many product items have an intermittent demand pattern that makes them all but impossible to forecast with traditional, smoothing-based forecasting methods. Items with intermittent demand – also known as lumpy, volatile, variable or unpredictable demand – have many zero or low volume values interspersed with random spikes of demand that are often many times larger than the average. This problem is especially prevalent in companies that manage large inventories of service and spare parts in industries such as aviation, aerospace, power and water supply and utilities, automotive, heavy asset management, high tech, as well as in MRO (Maintenance, Repair and Overhaul).
Smart Inventory Optimization use a unique empirical probabilistic forecasting approach that results in accurate forecasts of inventory requirements where demand is intermittent. The solution works particularly well whenever demand does not conform to a simple Normal distribution. Our APICS award-winning technology rapidly generates tens of thousands of possible scenarios of future demand sequences and cumulative demand values over an item’s lead time. These scenarios are statistically similar to the item’s observed data, and they capture the relevant details of intermittent demand without relying on the assumptions commonly made about the nature of demand distributions by traditional forecasting methods. The result is a highly accurate forecast of the entire distribution of cumulative demand over an item’s lead time. The bottom line is that with the information these demand distributions provide, companies can easily plan safety stock and service level inventory requirements for thousands of intermittently demanded items with nearly 100% accuracy.
Q: How is data imported/exported to and from other systems
GVH: Smart Software has developed out-of-the-box connectors with Epicor ERP, NetSuite, Dynamics AX, NAV, and 365 Business Central and are in the process of finalizing integrations to IBM Maximo and Epicor Prophet 21. We have integrated our platform with virtually every ERP solution on the market, including SAP, Oracle ERP, JD Edwards, PeopleSoft, Infor, Great Plains, Sage X3, and more using an import/export process supported by Smart Software’s data model and data processing engine. The data model and processing engine are designed to import and export data in a variety of file formats. Transaction data such as shipments, sales orders, supplier receipts are readily available in the ERP system. These are easily exported from the data source to a file and Smart IP&O takes care of the rest. All imported data is stored in our common data model, empowering automated analysis of trends, stocking policies, and inventory metrics. Forecasts and stocking policies are saved back to the ERP and can be updated “on demand” or via our scheduler.
Q. Smart IP&O is a SaaS solution. How do you ensure customer data is protected?
GVH: Great questions. First our software is hosted on AWS servers, not an unsecure “pizza box” somewhere. All kidding aside, my underlying point is that security of your data shouldn’t simply be assumed. We have undergone a very rigorous set of security and reliability audits that ensure Smart Software has established protocols and follows these protocols for handling customer data and ensuring security. Our software has been SOC Certified since 2018 and has met or exceeded the industry standards in all categories for reliability and security. In fact, our software is used by a nuclear power plant and the largest commuter railroad in the US to optimize parts inventory. They made sure our platform was secure before proceeding. Many software companies don’t bother to conduct these audits because they lack a formal process for handling your data and/or may not have the resources to incur the significant time and expense these audits entail.
Q: How much does your software cost and how is it sold?
GVH: Before I share details on price, it is important to emphasize the return on investment, as it dwarfs the upfront and ongoing investments. How much does it cost to not plan properly? To contend with stock-out after stock out, hits to on-time delivery, expedites, emergency production jobs, lost sales, and finger pointing? How much does it cost annually to hold excess inventory you don’t need? What about when that inventory needs to be written off when it becomes obsolete? During our sales process, we will review your data and together develop a credible estimate of bottom line impact. Many of our customers have saved millions.
Smart is sold as a subscription service only. The software is hosted on AWS data centers so you don’t have to worry about installs, databases, servers, networks, or upgrades. We take care of all of that for you as part of your subscription fee. Pricing is value-based and tied to your company’s (or business unit’s) revenue and inventory value. Pricing for small customers starts at $1,000/month, medium sized customers start at $3,500/month, and large customers at $10,000/month. Actual pricing is based on a number of factors including number of items you’ll manage, inventory value, revenues, and complexity of your business and planning process. We are happy to customize a quote based on the particulars of each situation.
Q. Do you offer a trial?
GVH: Yes. Smart is a SaaS company that embraces the “try before you buy” model for our prospective customers. This ensures you have a full understanding of what our software does and how it can be applied to your business before you committing to a full purchase.
Related Categories: