Our comprehensive solutions and services guide you on your digitalization journey.
Finding the right business solution:
We speak your language! Our industry-specific solutions are perfectly tailored to your needs.
Gartner® Report: Transform Your CRM with 5 Essential Steps
Tips and updates for our solutions, informative blog posts, compelling case studies, and news about our company.
The "Intelligent Automation" specialization is awarded exclusively to partners who meet strict criteria.
Our solutions can help you estimate your company's material and inventory requirements, as well as forecast future demand. This way, you can not only reduce internal supply chain planning efforts and avoid stock outs, but also optimise your cash flow.
Now that you have an overview of ALISA AI, let’s dive deeper into its four core functionalities:
Multiple (15+) algorithms forecasting demand based on historical data, 3rd party data and all relevant constraints on SKU level.
Algorithms define sales dynamics for each SKU, clean out-of-stock data and outliers in order to achieve baseline sales.
Forecast accuracy up to 97%
Same forecasting approach used to forecast quantities and be used to forecast financials, either revenue or cost side. Those forecasts can be utilized for top line annual planning, quarterly forecasts or monthly rolling forecasts.
A key differentiator in statistical forecasting is the integration of third-party data sets. Our platform includes weather history and weather forecasts by default. Additionally, based on industry specifics, we integrate various relevant third-party data sets, which can be structured or unstructured, such as global trends, supply chain updates, news on energy prices etc.
Forecasts on SKU-level are summed up to be used on store, region, country or any other level depending on how your supply chain is organized.
Different ordering rules can be set-up depending on if you have directy store distribution, central wh distribution or any other regional specific distribution format.
Integrate all types of historical promotion data and understand key drivers of sucessfull promotions. Also, integrate all upcoming promotions and automatically include those in ordering or replenishment plans.
Simulate potential impact promotion mix can deliver to your end result.
Ordering new season goods that have no historical sales can be supported by similar item detection algorithms.
Similar items can be detected by analyzing similar dimensions between items or by image recognition. Initial allocation is then done based on historical performance of most similar items.
During the season, platform can help replenish stores from central warehouse based on real-time performance of actual sales. Replenishment is done based on various logistical and store constraints.
Transferring goods from a non-performing to a performing store for each SKU is labor-intensive work. Our platform understands different sales dynamics on SKU (variant) and store level and provides you with a suggested transfer list weekly or/and other dynamics you define.
87% of transfers were sold within next 2 weeks
Final step in preparing vendor orders or transfers is trying to optimize transportation cost when possible. Cargo module takes into consideration all logistical information on item level and all perfomance infromation (revenu and margin) and comes up with proposal on how to optmially fill in truck or a container in order to achieve highest stock turnover ration.
Based on demand forecasts and based on currnetly open sales orders, our platform will provide you with suggested raw material and semi product requirements in order to fill in required production plans.
Also, those quantities will be converted automatically into suggested purchase or production orders with precise due dates taking all relevant constraints into production.
Even having all required materials and ingredients on-hand required complex production schedduling in order to full-fill all required due dates.
Our platform runs optimization algorithms which take all relevant constraints (machine, people and resouces) combine them with production recipes (BOM) and provides you with most optimal productions schedulle in order to maximize on-time fill rates.
Complex production organizations require workforce with various skillsets avaiable in various weekdays and hours.
Including all relevant constraints on your workforce into production schedduling is mandatory for receiving adequate workforce plan on weekly/monthly basis.
Integrating IoT data and reading relevant information from machines in productions is crucial input for predictive algorithms which can learn from historical performance and alert you on potential upcoming failover. This allows you to prevent failure, maintain the machine on-time, and thus reduce total failover time in which the machine is not operable.
Integrating camera on production lines and processing advanced computer vision algorithms over camera feed, provides customer with real-time anomally or defect detection and gives user instand infromation that certain action should be done with given item.
This real-time proccess has direct impact on improved and timely scrap handling.
Having all parameters important for modeling production, allows our platform to have a certain „digital twin” or a real production line. This allows us to simulate various scenarios when scheduling production and also allows you to easily and more reliably answer a question to your customer when can certainly delivery be done it also allows you to udnerstand what would be the impact of accepting a new work order and will it trigger any other late work orders or will it have negative revnue impact.
Platform front end is delivered through global leaders in self-service BI; Qlik Sense and Power BI. We decided to leverage the power of global leading products instead of developing customer visuals.
State of the art interactivity bring ease of use to a whole new level.
The key reason for utilizing the BI tool as a front-end is to visually explain forecast and recommendation results together with the historical performance of sales, purchases, or production.
This visual interpretaiton gives users confidence level into accepting recommendations provided by platform.
Platform includes custom made drag and drop interface together with input fields which allow end user to modify platforms recommendation for any output.
Those modifications are then saved and used again by machine learning algorithms in order to improve quality of the output in the next run.
The key reason for the high acceptance rate of our platform from the users' side is that their process is not changed but just augmented with high-quality information.
Platforms output which can be anything from order recommendation to production schedule for next week, is directly integrated into ERP and used from there to procees with a process in a same manner how it was done earlier. Platform doesn’t require users learning and using another system to create, modify or launch orders.
Next step in esing a technology to explain platforms result to an end user is using Gen AI to interpret platforms recommendations and key drivers of those recommendations, and explain those in conversational language via any desired chat platform.
Managing inventory? ALISA AI has the solution! In this video, discover how our platform effectively tackles inventory management challenges.
ALISA AI platform can predict demand with high confidence for up to 97% of the total items, taking off the burden from a purchasing department by using historical and 3rd party data.
Stockout or overstock risk is nullified by the daily stock monitor, one of the platform key components, which proactively tracks demand on every item, triggers and replans orders accordingly.
By constantly improving safety stock and keeping the monitor of market demand, our ALISA AI can help you avoid stock-out situations almost completely, lowering them by up to 90%.
In one to three months of productions, you'll be able to completely change the structure of your inventory, so that slow-moving items don't pile up anymore and your inventory is filled up with mid-to-high rotation items, which will help you improve inventory turnover for up to 2 times.
Not sure what you need? Contact us and we will be more than glad to help you.
Here’s how one of the leading distribution companies introduced AI into their supply chain and managed to avoid stockouts by using AI support.
Stock optimisation is an important KPI in supply chain businesses, as it translates to cash flow directly. On the one hand, constantly high stock levels (overstocking) indicate that money spent on stock is money not spent on potential growth while on the other hand, low stock levels (frequent stock-out events) indicate that money might be getting lost on missed sales.
Basic components in forecasting demand are historical data on market demand (retail sell-out data if possible), external 3rd party data, context, specifics, and various constraints which have significant impact on forecasting, such as promotion periods, price change, outliers, anomalies, availability, bonus schemes, etc. The results of a demand forecast are then used either to estimate the needed raw materials, or other manufacturing items in production companies, or just to define ordering goods for a vendor in trading companies.
There are many steps in a supply chain which can be optimised by using technology. Our solution focuses on helping companies order optimal quantities, which has an impact on multiple points in a supply chain, such as lower inventory stock, lower transportation stock, regular ordering frequencies, and better predictability, avoiding warehouse jams, etc.
Most of our customers experience benefits in a few business categories: financial benefits in terms of freed-up cash flow which can be invested into growth, labour benefits in terms of up to 50% less time spent on repetitive activities, which allows purchase departments to focus on complex purchases like new items placement, etc., as well as operational benefits in terms of lowering overstock levels while, at the same time, avoiding stock-out situations both in a warehouse and retail stores, or in raw materials acquisition.
On top of the benefits previously described, many other users within a company benefit from introducing AI based inventory optimisation. Marketing teams will have better insight into promotional efficiency, warehouse managers will be able to avoid warehouse jams due to predictive ordering periods, and your new employees or interns will gain knowledge organized in one place instead of it being scattered around the organization.
It is a service which estimates a demand for an individual item in your inventory, and generates an optimal order recommendation. It takes into account many constraints, specifics and context of an item, as well as the vendor.
Prior to entering a full project, we offer various piloting options ranging from a tailor-made presentation for your company to a detailed benefit analysis simulated on a digital twin using actual historical data and events. These POC activities also include workshops with key users for the purposes of discussing and validating the order recommendations generated by our algorithm.
As the first step of this journey, we offer analysis of your sales and stock data, and dedicated presentation of the results of our platform, including a tailor-made report for your company. Detailed steps and guidelines are listed here.
Dedicated workshops, called hypercare workshops, are an integral part of our projects where we sit together with end users and go through all the specifics that their vendors or SKU’s might have, and also help them to interpret the proposal coming out of the algorithms, as well as put them in the context of historical data and various constraints.
There are many business rules and constraints, dependant on the industry, that are a standard part of our solution like minimum order quantity or amount, seasonal sales, promotional sales (including various promotion types), cargo and logistics information, substitutes definition, leading items per category (prioritisation), etc.
Our solution can be deployed on-prem or in any private or public cloud environment, as well as API integrated with ERP.
