Review of INFORM Software, Supply Chain Optimization Software Vendor

Last updated: November, 2025

Go back to Market Research

INFORM Software (Inform Institut für Operations Research und Management GmbH) is a privately held German vendor founded in 1969 in Aachen that has grown into a mid-sized operations-research specialist with roughly one thousand employees and global subsidiaries. From the outset the company has focused on embedding mathematical optimization into operational decisions, initially in production planning and logistics and later expanding into aviation ground operations, finished vehicle logistics, risk & fraud analytics, workforce management, stocktaking, and supply chain management. Its portfolio is organized around several domain-focused products—ADDONE for inventory and supply chain planning, FELIOS for production planning and scheduling, GROUNDSTAR for airport and ground handling operations, and RISKSHIELD for fraud and risk scoring—rather than a single horizontal platform. Within supply chain, the ADDONE solution suite is marketed as “decision-intelligent optimization software” that creates a common data basis, generates AI-supported demand forecasts, proposes automated order quantities, and highlights exceptions for planners across demand planning, inventory management, spare parts management and S&OP. INFORM claims more than 1,000 customers worldwide across manufacturing, retail, logistics, finance and other sectors, and promotes case studies such as ARaymond and Hagebau Connect that attribute inventory reductions and process efficiency gains to ADD*ONE. Technically, the company combines operations research algorithms with machine learning – described as “decision intelligence” – implemented on a modern Java/Kotlin, Spring, SQL and microservices stack that can be deployed on web-based architectures and integrated with ERPs such as Microsoft Dynamics. At the same time, many technical details remain behind marketing language, so assessing how state-of-the-art INFORM really is in probabilistic forecasting, AI and large-scale optimization requires careful reading between the lines of its documentation, job postings and independent customer reports.

INFORM Software overview

From a supply chain perspective INFORM is best understood as a long-established operations-research house that has productized a set of vertical decision-support applications rather than as a general-purpose planning platform. The corporate entity – Inform Institut für Operations Research und Management GmbH – was founded in 1969 in Aachen by Hans-Jürgen Zimmermann, and has remained privately held with its headquarters and primary development center in Aachen.¹² Public figures for 2023 indicate roughly 968 employees and revenue around €129 million, placing INFORM as a sizeable but still mid-market vendor rather than a global mega-suite provider.¹ The firm emphasizes its operations-research heritage and describes itself as using “advanced mathematics for complex decisions,” positioning OR as the core of most of its systems and explicitly combining OR algorithms with machine learning techniques to improve decision quality in logistics, manufacturing, risk, and supply chain.²³

INFORM’s product line spans several distinct domains. Apps Run The World’s vendor profile and INFORM’s own marketing list at least eight named software products: ADDONE for inventory and supply chain management, FELIOS for production planning and scheduling, GROUNDSTAR for aviation ground operations, INVENT XPERT for stocktaking in SAP EWM, RISKSHIELD for fraud prevention and risk monitoring, SYNCROSUPPLY and SYNCROTESS for yard and transport scheduling, and WORKFORCEPLUS / YMSlite for workforce and yard management.⁴⁵³ Across these domains the unifying theme is algorithmic optimization of constrained, high-complexity processes. For supply chain, the ADDONE solution suite is the focal point: INFORM’s supply chain management pages explicitly frame ADD*ONE as “decision-intelligent optimization software” for demand planning, inventory management, spare parts planning, and S&OP, promising better service levels at lower inventories by automating routine planning tasks and surfacing critical exceptions.⁶⁷⁵

INFORM also positions itself within broader ESG and corporate-governance narratives, being registered as a participant in the UN Global Compact and highlighting sustainability initiatives in its corporate communications.⁸ This is relevant mostly as context: it signals that INFORM operates at a level of maturity where formal sustainability reporting and global compliance programs matter, rather than as a small, informal software shop. Industry directories such as It’s in Germany similarly describe INFORM as an AI-based optimization vendor with more than 1,000 employees serving international customers in logistics, automotive, aviation and finance, reinforcing the view that INFORM is an established European mid-sized vendor with diversified operations rather than an early-stage startup.⁹

Company history, ownership and scale

The German Wikipedia entry and INFORM’s own “Our history” page converge on a coherent timeline. INFORM began in 1969 as an academic spin-out in operations research, developing software to support complex decision problems in manufacturing and logistics.¹² Over the following decades it added vertical product lines – notably for aviation ground operations (GROUNDSTAR), production planning (FELIOS), and container terminal logistics – and geographically expanded with subsidiaries in North America, South America, Asia and Australia.²⁴ As of the mid-2020s INFORM reports more than 1,000 active customers across 40+ countries, with a workforce approaching 1,000 people.²¹⁰⁴

There is no public evidence of INFORM having been acquired or itself acquiring other software companies. Corporate registries and industry news items do not show M&A activity associated with the Aachen-based INFORM entity; one has to be careful not to confuse it with unrelated firms that share the word “Inform” in their names (for example, an insurance analytics firm in the US that was acquired by Klear.ai, which is a different company). In other words, INFORM appears to have grown organically as a privately held, founder-influenced company focused on operations-research-based products, rather than through venture-backed consolidation or acquisition roll-ups.¹⁴

Product portfolio with focus on supply chain

The ADDONE suite is marketed as INFORM’s dedicated supply chain and inventory optimization product line. INFORM’s supply chain management page describes ADDONE as “software for supply chain management” that supports demand planning, inventory management, spare parts management and S&OP, with a promise to “reduce inventories, increase service levels and improve planning quality” using AI-supported software.⁶ The demand-planning subpage specifies that ADD*ONE acts as “sales forecasting software” combining “field-tested optimization algorithms and artificial intelligence” to generate reliable forecasts and to support sales planners in aligning forecasts with sales targets, campaign plans and strategic objectives.⁷

On the S&OP side, the ADD*ONE-based S&OP page emphasizes the creation of a common information base across departments, automatic generation of forecasts and “self-adapting demand indicators,” and provision of dashboards plus decision support for balancing supply, demand and capacity over 24-month horizons.⁵ The same page lists the main benefits as improving cross-departmental planning quality, networking departmental knowledge, optimizing inventory levels and capacity utilization according to demand, and enabling “management by exception” by proactively flagging critical items where planner intervention is required.⁵

Case material gives more concrete flavor to these claims. A success story with ARaymond, a global fastening and assembly solutions supplier, states that ADD*ONE forms “the essential data basis for our now-established S&OP process, all from one system,” enabling simplification and improved robustness of the S&OP process with better transparency over the end-to-end plan.¹¹ Another independent report from Retail Optimiser describes how Hagebau Connect, the e-commerce unit of German building materials group Hagebau, implemented the web-based Add-One solution for e-commerce replenishment, automated key procurement processes and – together with other measures – reduced inventory by 30% within the first six months.¹² According to that article, the solution integrates with Microsoft Business Central via an interface, uses AI-based forecasting procedures to generate automated order proposals and presents planners with tailored workflows and visual aids such as stock coverage graphs and supplier performance indicators.¹²

These customer stories support the picture that ADD*ONE is not just a reporting or alerting layer, but an optimization engine that generates concrete replenishment proposals and S&OP plans which customers can execute. However, the public material remains largely descriptive and does not disclose detailed modeling assumptions (for example, whether demand and lead times are modeled as full probability distributions or via simpler safety-stock parameters), so an assessment of technical depth must rely on indirect evidence such as the OR and AI pages and technical job postings.

INFORM Software vs Lokad

Although both INFORM and Lokad target supply-chain decision problems and invoke AI and advanced analytics, they embody quite different architectural and commercial philosophies. INFORM offers a catalog of domain-specific standard products – ADD*ONE for inventory and supply chain, FELIOS for production planning, GROUNDSTAR for airport operations, RISKSHIELD for fraud, and others – each with its own UI, workflows and embedded optimization logic.⁵⁴³ Lokad, by contrast, exposes a single programmable platform centered on its Envision domain-specific language, where all forecasting, optimization and reporting logic are implemented in code and tailored per client. Lokad’s deliverable is typically a bespoke predictive-optimization app built on this platform for a client’s supply chain, while INFORM aims to sell configurable but largely pre-packaged applications for specific problem domains.

On the forecasting side, INFORM’s supply chain pages repeatedly mention “AI-supported software,” “intelligent forecasting procedures” and “self-adapting demand indicators” that generate reliable forecasts and feed automated replenishment proposals.⁶⁷⁵¹² The descriptions are consistent with statistically or machine-learning-enhanced time-series forecasting, yet they are silent on whether the system produces full demand distributions or a limited set of point forecasts and safety stock parameters. Lokad, in contrast, explicitly centers probabilistic forecasting and quantile distributions as the technical backbone of its optimization; its M5 competition performance and public documentation show that its forecast outputs are treated as full stochastic inputs into downstream decision models. This does not mean INFORM lacks stochastic modeling – its operations-research page notes that OR is combined with machine learning predictions to explore large decision spaces – but from public sources it is not possible to confirm that ADD*ONE’s demand modeling is fully distributional rather than using more traditional safety-stock-style buffers.³

In optimization, INFORM foregrounds operations research and mathematical programming. Its OR page describes the use of mathematical modeling, optimization and search algorithms to explore huge decision spaces and quickly converge on “optimal solutions,” and explicitly states that many INFORM systems (including in logistics, manufacturing and workforce) are based at their core on OR search and optimization algorithms.³ ADD*ONE is presented as “decision-intelligent optimization software” that can automate routine replenishment decisions, allocate limited resources across products and time, and highlight exceptions for human attention.⁶⁵ Lokad, meanwhile, uses custom stochastic-optimization heuristics such as Stochastic Discrete Descent and introduces differentiable programming to jointly learn forecasting and decision parameters; it deliberately eschews off-the-shelf optimization solvers in favor of domain-specific search procedures embedded in its DSL. Technically, both vendors are in the optimization business, but INFORM leans on classical OR terminology and packaged optimization engines within fixed product boundaries, while Lokad exposes optimization as code in a unified environment.

Architecturally and technologically, INFORM’s own job postings and third-party descriptions show a relatively conventional enterprise stack: backend development in Java or Kotlin with Spring and SQL, use of Git, Maven and Jira, and deployment via modern cloud technologies such as Docker, Kubernetes, microservices, domain-driven design and messaging systems, with AWS cited as an example environment.¹³ This suggests a microservices-based architecture backed by relational databases and web front-ends per product line. Lokad, by contrast, is built on a custom event-sourced architecture with a content-addressable store and a distributed VM for executing Envision scripts, implemented in F# and C# on Microsoft Azure. From a customer’s perspective, INFORM’s advantage is familiarity – its solutions look and behave like contemporary enterprise applications with integrated UIs and standard interfaces to ERPs such as Microsoft Dynamics and SAP – while Lokad’s advantage is programmability and the ability to model highly idiosyncratic constraints at the expense of requiring scripting and closer collaboration.

Commercially, INFORM is a long-standing mid-sized vendor with more than five decades of history and a broad portfolio across industries; its supply chain offering is one important line of business among many.¹⁴ Lokad is a much younger, narrower vendor whose entire business is focused on quantitative supply chain optimization. INFORM’s breadth – aviation, logistics, risk & fraud, workforce – brings diversification and cross-domain OR expertise, but it also means that supply chain is not the company’s sole strategic focus. Lokad’s depth in supply chain translates into a platform that is more specialized but less “out-of-the-box” for other domains. For a buyer, the trade-off is between INFORM’s productized, domain-specific applications with OR/AI embedded in each, and Lokad’s single programmable engine that demands more configuration work but can, in principle, express more sophisticated probabilistic and economic models tailored to each business.

Technology and architecture

Tech stack and deployment

INFORM does not publish a formal architecture blueprint for ADDONE, but its career materials and partner write-ups provide a reasonable picture. A Software Developer job posting in Aachen specifies that developers “extend our software solutions in the area of container terminals with Java or Kotlin,” work on the “cloud transformation of our product,” and are expected to have “very good knowledge in Java or Kotlin, Spring and SQL” plus experience with Git, Maven and Jira.¹³ The same posting lists experience with CI/CD, Docker, Kubernetes, microservices, domain-driven design, messaging systems and cloud platforms such as AWS as advantageous, and mentions Angular on the front-end.¹³ While this posting references logistics software for container terminals, INFORM’s hiring patterns strongly suggest that the same core technologies are used across multiple product lines, including ADDONE.

Third-party vendor directories similarly describe INFORM’s products as web-based or cloud-enabled solutions. The Retail Optimiser article on Hagebau Connect explicitly calls ADD-ONE a “web-based solution” deployed via browser-based interfaces and integrated through an interface to Microsoft Business Central, with workflows that present users with relevant information in a browser-based UI.¹² Apps Run The World’s vendor profile notes INFORM as an AI and decision-intelligence software vendor delivering its products to more than 1,000 customers worldwide, with a mix of cloud and on-premises deployments depending on product and customer preference.⁴ It’s in Germany likewise characterizes INFORM as offering AI-based software solutions that can be integrated into existing IT landscapes, further reinforcing the view of a fairly standard web/app-server/DB architecture.⁹

From this evidence, a plausible architecture for ADD*ONE is:

• A multi-tier web application, with Angular or similar SPA front-ends and Java/Kotlin Spring-based services on the backend.¹³¹²
• Relational databases (SQL) storing transactional and master data loaded from ERP/WMS/other systems.
• Microservices and messaging for scaling and integration, with optional containerization (Docker, Kubernetes) and cloud deployment (AWS or private clouds).¹³
• Embedded optimization engines and forecasting modules implemented as service components within the ADD*ONE suite rather than as separate, externally orchestrated pipelines.⁶⁷³

This is a modern but conventional enterprise stack: the technical interest lies less in the infrastructure than in the optimization and forecasting algorithms embedded in the services.

Data integration and roll-out

INFORM positions ADDONE as an overlay that integrates with existing ERPs and operational systems via standard interfaces. In the Hagebau Connect case, the Retail Optimiser article notes that the software is connected to Microsoft Business Central (formerly Navision / Dynamics NAV) through an interface that transfers all relevant data into ADD-ONE, after which six planners and two process managers control around 3,500 active SKUs using the solution.¹² According to the same source, the goal was to move from manual Excel-based quantity planning and manual ordering in the ERP to automated replenishment proposals generated daily by ADDONE, with workflows and visualizations that structure the planners’ daily work.¹²

INFORM’s own supply chain pages also stress that ADDONE creates a “common database” across departments as a basis for collaborative planning, and that it can integrate strategic decisions with operational plans over up to 24 months.⁵ While the exact integration mechanisms (file-based, API, message queues) are not disclosed publicly, the examples suggest a classic nightly or periodic batch process: ERP and sales data are extracted into ADDONE, calculations are performed, and replenishment proposals and dashboards are generated for planner review.

There is no detailed public description of implementation methodology, but the ARaymond case implies a project approach with phases for data integration, configuration, and iterative adaptation. ADD*ONE was described there as providing a unified data basis and simplifying the S&OP process once established, implying a non-trivial setup and change-management effort before benefits materialize.¹¹ Compared to a programmable platform like Lokad’s, INFORM’s roll-out hinges more on configuring an existing application and less on writing domain-specific code, but in both cases success depends on the quality of data integration and the tuning of models and parameters.

AI, machine learning and optimization claims

Operations research core

INFORM’s “Operations Research – Mathematical Optimization” page offers a rare look at the conceptual underpinnings of its optimization engines.³ It defines OR as using sophisticated analytical methods and mathematical algorithms to make better decisions in complex situations, describing how mathematical modeling and optimization can represent the entire decision space of a problem with objectives and constraints, and how OR algorithms search this space to find the best decisions in a short time.³ The page explicitly states that many of INFORM’s software systems are based at their core on OR search and optimization algorithms, particularly in logistics, manufacturing and workforce roster generation, and that OR is combined with machine-learning-based predictions to optimize business processes.³

While no specific algorithms are named (e.g., mixed-integer programming, constraint programming, heuristics), the language is consistent with classic OR practice: formulating a mathematical model and using either exact or heuristic algorithms to explore it. Given the diversity of INFORM’s domains (container terminals, assembly lines, airport resource allocation, etc.), it is reasonable to infer that a mix of exact and heuristic methods are used, but this remains an inference – the company does not publish technical whitepapers or open-source solver code that would allow a deeper audit. The presence of such an OR-focused page, however, and the company’s long history in OR, support the claim that optimization is more than a superficial marketing label.

AI and machine learning

INFORM uses the label “decision-intelligent” and repeatedly references artificial intelligence in its marketing. The supply chain management page describes ADDONE as “AI-supported software” that leverages AI to optimize inventory levels, delivery performance and capacity utilization, and the demand-planning page speaks of “field-tested optimization algorithms and artificial intelligence” for forecasting.⁶⁷ The S&OP page notes that ADDONE “uses artificial intelligence methods to optimize your processes” and automatically provides reliable forecasts and “self-adapting demand indicators.”⁵

Blog posts linked from the S&OP page, such as “Inventory Optimization with AI-supported Software: How to Reconcile Delivery Capability and Cost Reduction” and “Supply Chain Management Software: Why it is Essential for Successful Supply Chains,” further promote AI capabilities in inventory optimization and supply chain resilience.⁵ However, these blog articles (as far as can be seen from snippets and titles) remain at a conceptual and business-explanation level; they do not disclose model architectures, feature engineering approaches or training procedures.

The Hagebau Connect case from Retail Optimiser is more concrete: it calls the solution “AI-based” and explains that “intelligent forecasting procedures” analyze demand behavior for each article and create automated order proposals, which are then presented in a workflow with visualizations like coverage charts and supplier quality indicators.¹² Yet again, the actual AI models are not specified. It is plausible that INFORM uses a combination of classical time-series models and machine-learning methods (e.g. gradient-boosted trees, neural networks) within ADD*ONE, but this cannot be confirmed from public material.

From a skeptical standpoint, the AI claims are credible in the sense that INFORM clearly uses data-driven models for forecasting and event detection, and hires developers with experience in “AI software projects” as a plus.¹³ But the lack of technical disclosure means one cannot verify whether the AI is genuinely at the level of state-of-the-art probabilistic forecasting and decision learning (as in modern academic literature) or mainly a combination of solid but conventional statistical models wrapped in AI-marketing language. There is no evidence that INFORM publishes in forecasting competitions or AI conferences, nor that it exposes model internals for customer inspection beyond standard dashboards.

Optimization and automation in ADD*ONE supply chain

In supply chain, the key question is how far ADDONE goes beyond basic reorder-point logic. INFORM’s pages describe ADDONE as generating reliable forecasts, calculating optimal inventory levels, and automatically providing replenishment proposals that are optimized for availability and cost, with the system taking over “tedious routine tasks” and proactively alerting planners to critical items.⁶⁵ The S&OP material emphasizes that plans are “actionable, cross-departmental” and that the software can optimally allocate limited resources and adjust business plans tactically and strategically over a 24-month horizon.⁵

The Hagebau Connect article gives evidence of real automation: prior to ADDONE, Hagebau Connect’s planners relied on manual Excel analyses and manually triggered orders in the ERP; after implementation, central procurement processes were automated, planning frequency increased from once or twice per week to daily, and (together with other measures) the inventory level dropped by 30%.¹² The article also notes the use of rule-based calendars, reachability graphs and supplier performance metrics within the ADDONE UI to support decision-making.¹²

This suggests that ADDONE’s optimization is at least at the level of dynamic order proposal generation that considers demand forecasts, stock levels, open orders and possibly supplier constraints, and that it supports exception-based planning. However, because INFORM does not publish the underlying formulas or objective functions, we cannot determine whether the optimization is based on simple reorder-point calculations with some heuristics, or on full stochastic optimization of expected costs under demand and lead-time uncertainty. The operations-research page implies the latter for some domains, but no explicit link is made to ADDONE’s inventory logic.³

Overall, INFORM’s optimization claims are credible and consistent with long-standing OR practice in APS systems: they likely represent solid, industrial-strength optimization models embedded in standard software, but there is insufficient evidence to assert that they are ahead of the state of the art in probabilistic inventory optimization.

Client base and commercial maturity

Named clients and case evidence

INFORM provides a references section and multiple case stories on its site, though many are behind forms or summarized in short blurbs. The ARaymond S&OP story, for example, describes how ADD*ONE supports ARaymond (a global fastening technology company active in automotive and aviation) by providing a single data basis for the S&OP process, simplifying workflows, and underpinning more efficient cross-functional planning; the quoted team lead states that the process is now “based on a valid foundation and therefore more efficient.”¹¹ The sector here – industrial manufacturing supplying automotive and aviation – is consistent with INFORM’s focus on complex, engineer-to-order and long-tail inventory structures.

The Hagebau Connect case discussed above is particularly informative because it comes from an independent trade publication. Retail Optimiser reports that using ADD-ONE, Hagebau Connect automated e-commerce replenishment, integrated Microsoft Business Central and, after six months, reduced stock levels by 30% while enabling daily planning thanks to the elimination of manual steps.¹² The article gives concrete numbers (3,500 SKUs, six planners and two process managers) and describes how the system presents planners with prioritized information, which strengthens its evidentiary value.

Apps Run The World’s profile lists INFORM as serving more than 1,000 customers across industries including automotive, aviation, logistics, manufacturing and financial services, and highlights products such as ADD*ONE, FELIOS, GROUNDSTAR and RISKSHIELD as key offerings used by these clients.⁴ While such directories rely partly on vendor-provided data, they do corroborate INFORM’s claims of scale and cross-industry adoption.

Overall, compared to many early-stage AI planning startups, INFORM has a substantial and verifiable customer base with decades-long deployment history in multiple domains. For supply chain specifically, the public case material is more limited but still includes named, recognizable industrial clients with documented benefits.

Market positioning in supply chain

INFORM’s supply chain positioning is narrower than that of global planning suites like SAP IBP or Blue Yonder, but broader than niche players focused solely on demand forecasting or single-echelon inventory optimization. The ADD*ONE suite covers demand planning, inventory management, spare parts management and S&OP; in addition, the FELIOS and stocktaking products can be relevant to production planning and inventory audits in SAP environments.⁶⁵⁴ This places INFORM squarely in the “optimization-centric APS” category: it provides focused, optimization-heavy applications that integrate with ERP backbones rather than full end-to-end transactional systems.

Industry write-ups such as those on Retail Optimiser frequently mention INFORM alongside other European supply-chain IT vendors in the context of grocery, DIY and wholesale logistics, indicating that it competes as one of several options for mid- to large-sized retailers and manufacturers looking to automate replenishment and planning.¹² It’s in Germany and similar directories emphasize its AI-based decision-intelligence positioning, but do not rank it as a top-tier global player; rather, INFORM appears as a strong regional/European contender with particular strengths in Germany and adjacent markets.⁴⁹

Commercially, INFORM is clearly a mature company: five decades old, near-€130m revenue, 1,000+ employees, broad industry coverage and participation in global governance frameworks.¹²⁸ On the supply-chain-technology curve, this places it in the “established vendor” category: not a legacy mainframe-era suite, but also not an experimental AI startup. The main uncertainty is not about commercial viability but about how aggressively its ADD*ONE suite has evolved to incorporate the latest probabilistic and AI techniques versus iterating on traditional OR-based APS models.

Critical assessment of state-of-the-art

Putting the pieces together, INFORM’s technical profile in supply chain looks like this:

• Strengths

  • Deep roots in operations research with explicit emphasis on mathematical optimization across products, not just in marketing.²³
  • A modern enterprise tech stack (Java/Kotlin, Spring, SQL, microservices, Docker/Kubernetes, cloud) that should be maintainable and scalable for typical planning workloads.¹³
  • A productized suite (ADD*ONE) that covers multiple key supply chain planning functions – demand planning, inventory optimization, spare parts and S&OP – in an integrated way.⁶⁷⁵
  • Evidence of real-world automation and impact, such as the Hagebau Connect case with 30% inventory reduction and daily e-commerce replenishment, and ARaymond’s S&OP process simplification.¹¹¹²
  • A sizeable and diversified customer base across industries, signaling robustness and long-term viability.⁴⁹

• Ambiguities / limitations

  • Limited transparency on forecasting models. Public material does not clarify whether ADD*ONE models full probability distributions for demand and lead time, or mainly supports point forecasts plus safety stocks. Phrases like “reliable forecasts” and “self-adapting demand indicators” are marketing-friendly but technically vague.⁷⁵¹²
  • Lack of technical publications. Unlike some vendors that publish in forecasting competitions or academic venues, INFORM does not appear to provide technical whitepapers or benchmark results that would allow rigorous comparison of forecasting or optimization performance.
  • Black-box optimization from the user’s standpoint. While OR is central under the hood, the optimization logic is encapsulated in the product rather than exposed as configurable code. Users can adjust parameters and policies, but there is no evidence that they can fundamentally reshape the optimization model as they could in a programmable platform.
  • Conventional architecture. The Java/Spring/SQL/microservices stack is robust and industry-standard but not particularly innovative in itself; the innovation resides (if anywhere) in the algorithms and OR models, which are not disclosed.¹³
  • Broad focus diluting supply-chain R&D. INFORM’s product and R&D capacity is spread across aviation, logistics, fraud, workforce and other domains as well as supply chain; supply-chain optimization is one important line, but not the firm’s sole technological frontier.

Relative to the state of the art in supply-chain analytics – characterized by probabilistic forecasting, Monte Carlo-based stochastic optimization, and, in some research, differentiable programming that jointly optimizes forecasts and decisions – INFORM’s public positioning suggests it is at least keeping pace conceptually (AI, OR, decision intelligence) but does not provide enough technical evidence to conclude that it is at the leading edge. It is more accurate to view INFORM as a mature OR-based vendor that has incorporated modern machine learning where useful, wrapped in a decision-intelligence narrative, and successfully industrialized this in product form.

This is not a criticism of effectiveness: many supply chains would see substantial gains simply from adopting the level of automation and OR that INFORM demonstrably provides, as illustrated by cases like Hagebau Connect.¹² The skepticism applies to marketing claims around AI and “optimal solutions” – without algorithmic detail or independent benchmarks, these claims should be interpreted as “solid industrial optimization” rather than cutting-edge AI research deployed at scale.

Conclusion

INFORM Software is a technically credible, commercially mature vendor whose core competence lies in embedding operations research into domain-specific applications across logistics, production, risk and supply chain. In the supply-chain arena, its ADD*ONE suite offers an integrated package for demand planning, inventory management, spare parts planning and S&OP that can demonstrably automate replenishment, support exception-based planning and deliver significant inventory and process benefits for customers such as ARaymond and Hagebau Connect.¹¹¹² The underlying technology stack – Java/Kotlin, Spring, SQL, microservices, Docker/Kubernetes – is modern and industry-standard, and the company’s five-decade OR heritage and global customer base give it substantial practical experience.¹²⁴¹³

At the same time, INFORM’s public documentation and marketing materials reveal relatively little about the precise forecasting and optimization algorithms that power ADD*ONE, beyond high-level references to OR and AI. This opacity is typical of many commercial APS vendors, but it limits the ability of external observers to assess whether INTRO’s models implement fully probabilistic, economically grounded optimization or more traditional safety-stock and heuristic approaches framed in contemporary language. Compared with a programmable platform like Lokad’s, INFORM offers less transparency and flexibility but more productized, ready-made applications – a trade-off that may be attractive for organizations seeking packaged software with configurable parameters rather than custom-coded optimization pipelines.

In short, INFORM should be seen as an OR-heavy, mid-market planning vendor with robust, field-proven applications rather than as an avant-garde AI research outfit. For companies with conventional ERP backbones and a desire to automate replenishment and S&OP using established best practices in OR and time-series forecasting, ADD*ONE appears capable of delivering real value, as long as expectations about “AI” and “decision intelligence” are grounded in the reality of well-engineered but largely opaque optimization engines rather than in visions of fully self-learning, end-to-end autonomous supply chains.

Sources