DCT
1:25-cv-00516
Chep Technology Pty Ltd v. Ivisys Ab
Key Events
Amended Complaint
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: CHEP Technology Pty Limited (Australia) and CHEP Ltd (New York)
- Defendant: Ivisys Ab (Sweden), IVISYS Sweden AB (Sweden), and Ivisys Inc. (Delaware)
- Plaintiff’s Counsel: Morris, Nichols, Arsht & Tunnell LLP
- Case Identification: 1:25-cv-00516, D. Del., 11/07/2025
- Venue Allegations: Venue is alleged to be proper in the District of Delaware because Defendant Ivisys Inc. is a Delaware corporation registered in the district.
- Core Dispute: Plaintiff alleges that Defendant’s PalletAI automated pallet inspection system infringes three U.S. patents related to the automated inspection and repair of commercial shipping pallets.
- Technical Context: The technology involves industrial machine vision systems, using lasers and cameras to create three-dimensional models of objects to assess their condition and automate quality control in logistics and supply chain management.
- Key Procedural History: The complaint alleges that Plaintiff sent a notice letter in December 2022 to Alliance Automation, LLC ("Alliance"), Defendant's eventual exclusive North American partner, regarding a similar system. In February 2023, Alliance’s CEO allegedly acknowledged that its system "does seem to infringe upon the subject patent." Defendant and Alliance subsequently announced their partnership in December 2023 to commercialize the accused PalletAI system. Defendant was served with the original complaint in this action in May 2025.
Case Timeline
| Date | Event |
|---|---|
| 2002-12-10 | Earliest Priority Date (’668, ’976 Patents) |
| 2003-12-19 | Earliest Priority Date (’360 Patent) |
| 2010-08-03 | '668 Patent Issued |
| 2014-11-11 | '360 Patent Issued |
| 2014-12-30 | '976 Patent Issued |
| 2022-12-20 | Plaintiff sends notice letter to Alliance regarding alleged infringement by iPallet system |
| 2023-02-20 | Alliance CEO allegedly acknowledges infringement of a CHEP patent |
| 2023-12-01 | Defendant IVISYS and Alliance announce exclusive partnership to commercialize PalletAI system |
| 2024-03-23 | IVISYS announces order for four PalletAI systems to iGPS, delivered with Alliance |
| 2025-05-14 | Original complaint served on Defendant IVISYS, INC. |
| 2025-11-07 | First Amended Complaint Filed |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 7,765,668 - "Automated Pallet Inspection and Repair" (Issued Aug. 3, 2010)
The Invention Explained
- Problem Addressed: The patent’s background section describes the traditional repair of pallets as a manual process requiring handling and inspection by an operator (U.S. Patent No. 7,765,668, col. 2:30-36). This manual process is described in the complaint as tedious, costly, prone to human error, and raising safety concerns (Compl. ¶25).
- The Patented Solution: The invention proposes an automated work cell where a pallet is gripped and transported by a robotic arm through an inspection station (Compl. ¶37; ’668 Patent, col. 2:46-51). A sensing head, such as a laser and camera system, creates a three-dimensional data map of the pallet to detect defects like gaps or protrusions (’668 Patent, col. 4:60-66). This 3D data is then filtered into a 2D image, and a processor generates a "recipe of repair tasks" that is sent to automated repair tools (’668 Patent, col. 5:1-6:10). This process allows for consistent, automated analysis and directs subsequent robotic repair operations (’668 Patent, col. 3:5-19).
- Technical Importance: The invention aimed to replace subjective, manual inspection with an objective, automated process, thereby increasing the consistency, speed, and safety of pallet sorting and repair in high-volume logistics operations (Compl. ¶32).
Key Claims at a Glance
- The complaint asserts independent method Claim 3 (Compl. ¶95).
- Essential elements of Claim 3 include:
- using a scanning device to create a three-dimensional data map of a pallet for detecting gaps and protrusions in the pallet;
- filtering the three-dimensional data map into a two-dimensional image of on/off values by using a dynamically created height value, corresponding to a reference plane or set threshold offset above a board surface of the pallet;
- creating a recipe of repair operations from the three-dimensional data map; and
- transporting the pallet to at least one repair station in accordance with the recipe.
- The complaint does not explicitly reserve the right to assert other claims of the '668 Patent.
U.S. Patent No. 8,881,360 - "Software and Methods For Automated Pallet Inspection and Repair" (Issued Nov. 11, 2014)
The Invention Explained
- Problem Addressed: The patent identifies a need for a process that not only determines if a pallet needs repair but also generates a specific "list of repairs to be made" for an automated repair station ('360 Patent, col. 2:18-22). This contrasts with prior systems that made only a "pass-fail decision" without providing detailed repair instructions (Compl. ¶50).
- The Patented Solution: The invention claims a software-driven method for operating an inspection station. A laser and camera capture 3D coordinate data from the pallet ('360 Patent, col. 2:50-53). A computer system then analyzes this data through a multi-step process: it filters the 3D points to isolate the top surface, identifies individual elements (like boards and bearers), determines the overall pallet design based on those elements, and compares each element against design criteria in a database to generate a list of required repairs ('360 Patent, col. 2:58-67). Figure 4 of the patent provides a logic flow chart detailing how gaps between boards are analyzed to determine if a repair is needed ('360 Patent, Fig. 4).
- Technical Importance: This technology provided a more sophisticated analytical framework, enabling automated systems not just to sort pallets but to generate the specific instructions required for fully automated repair, a key step in removing human decision-making from the process (Compl. ¶50).
Key Claims at a Glance
- The complaint asserts independent method Claim 9 (Compl. ¶170).
- Essential elements of Claim 9 include:
- operating at least one laser to direct light toward a pallet;
- operating at least one camera to collect the reflected light and generate three-dimensional points (with x, y, and z coordinates); and
- analyzing the reflected light by: (i) filtering the points to obtain a top surface by discarding points with a z-coordinate below a threshold, (ii) identifying the type and number of each element, (iii) determining a pallet design from the elements, and (iv) comparing each element to a database to generate a list of repairs.
- The complaint does not explicitly reserve the right to assert other claims of the '360 Patent.
U.S. Patent No. 8,918,976 - "Automated Digital Inspection and Associated Methods" (Issued Dec. 30, 2014)
- Technology Synopsis: This patent claims a physical system (an "automated pallet inspection station") rather than a method. It addresses the same problem of automating pallet inspection by claiming the hardware combination of a frame, a transport system (e.g., a conveyor) carried by the frame, at least one "pallet feature sensing head" that generates a 3D data map, and a filter that processes the 3D map into a 2D image (Compl. ¶54-57).
- Asserted Claims: The complaint asserts independent system Claim 1 (Compl. ¶261).
- Accused Features: The complaint alleges that the PalletAI system's physical structure, including its frame, conveyor belts, and its integrated camera and laser assembly, infringes the claimed system (Compl. ¶269, ¶271, ¶275).
III. The Accused Instrumentality
Product Identification
The accused instrumentality is the "PalletAI pallet inspection system," which the complaint alleges was formerly known as "PalletInspector" (Compl. ¶76-77).
Functionality and Market Context
The complaint, citing Defendant's website, describes the PalletAI system as an inspection unit that uses a combination of cameras, lasers, and tailored illumination to scan pallets (Compl. ¶78). It is alleged to be capable of inspecting attributes such as overall dimensions, individual board dimensions, and defects like cracks, missing boards, or protruding nails (Compl. ¶78). The system allegedly includes a graphical user interface (GUI) that displays 2D images of the scanned pallet and can be integrated with a sorting system to route pallets based on the inspection results (Compl. ¶79). The complaint alleges the PalletAI system is marketed and sold in North America through an exclusive partnership with Alliance Automation, with installations at pallet recycling and manufacturing companies (Compl. ¶80, ¶82, ¶83). The complaint presents a screenshot of the PalletAI system in operation on a production line, showing a pallet moving on a conveyor through a large inspection housing (Compl. ¶101).
IV. Analysis of Infringement Allegations
7,765,668 Patent Infringement Allegations
| Claim Element (from Independent Claim 3) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| using a scanning device to create a three-dimensional data map of a pallet for detecting gaps and protrusions in the pallet | The PalletAI system uses a machine vision system with cameras and lasers to collect 3D data and determine the "dimensional accuracy" of pallets in length, width, and height. | ¶105, ¶111 | col. 4:60-66 |
| filtering the three-dimensional data map into a two-dimensional image of on/off values by using a dynamically created height value, corresponding to a reference plane or set threshold offset above a board surface of the pallet | The PalletAI system transforms the collected 3D data map into a 2D image, which is depicted on a GUI display screen. | ¶116, ¶119-120 | col. 5:1-9 |
| creating a recipe of repair operations from the three-dimensional data map | The PalletAI system is allegedly part of an automated "sort and repair system" and is capable of identifying repair requirements. | ¶124-126 | col. 6:1-5 |
| and transporting the pallet to at least one repair station in accordance with the recipe. | The system allegedly can instruct a conveyor to move damaged pallets to specific repair stations based on the defects identified during scanning and data processing. | ¶128-130 | col. 6:10-19 |
Identified Points of Contention
- Scope Questions: A central question may be whether the PalletAI system's output constitutes a "recipe of repair operations" as required by the claim. The defense may argue that generating a sort command or a list of defects is distinct from creating a "recipe" which the patent describes as controlling a "machine control within the repair cell" ('668 Patent, col. 6:6-8).
- Technical Questions: The complaint alleges the system "transforms" 3D data into a 2D image "by using a dynamically created height value" (Compl. ¶122). A factual question will be what evidence supports the allegation that the accused system uses this specific filtering technique, as opposed to other methods of generating a 2D representation from 3D data. The complaint includes an image from Defendant's website showing a 3D point cloud representation of a pallet, which may support the "three-dimensional data map" element (Compl. ¶114).
8,881,360 Patent Infringement Allegations
| Claim Element (from Independent Claim 9) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| operating at least one laser to direct light toward a pallet comprising a number of elements to be inspected | The PalletAI system is advertised as using a combination of cameras, lasers, and tailored illumination to inspect various pallet elements. | ¶179-180 | col. 2:50-53 |
| operating at least one camera to collect the light reflected from the pallet, and generating three-dimensional points based on the reflected light, with each point having an x-, a y-, and a z-coordinate | The PalletAI system allegedly uses a machine vision system with at least one camera to collect 3D data on pallets, generating three-dimensional points to detect dimensional accuracy. | ¶182-183, ¶189 | col. 3:25-30 |
| analyzing the reflected light based on... (i) filtering the three-dimensional points to obtain a top surface geometry and topography by discarding points having a z-coordinate below a threshold | The complaint alleges on information and belief that the system transforms a 3D data map to obtain top surface geometry by discarding points with a z-coordinate below a threshold. | ¶201, ¶208 | col. 3:30-40 |
| (ii) identifying a type and number of each element based on the top surface geometry and topography | The system allegedly analyzes the top surface of pallets, including individual boards, to identify elements. | ¶210, ¶212 | col. 2:58-61 |
| (iii) determining a pallet design from the type and number of each element | The system allegedly detects the type of pallet and displays it on the GUI. | ¶214, ¶216 | col. 2:61-62 |
| and (iv) comparing each element of the pallet to a database to generate a list of repairs. | The system allegedly compares inspected pallet elements to a set of parameters or thresholds to determine if repair is needed and can generate a report listing defects. | ¶218, ¶220 | col. 2:62-3:2 |
Identified Points of Contention
- Scope Questions: The interpretation of "determining a pallet design" will be significant. The defense could argue that simply identifying a pre-programmed pallet type (e.g., "GMA 48x40") is not the same as "determining a pallet design from the type and number of each element" as taught by the patent, which suggests a more foundational analysis of the pallet's structure.
- Technical Questions: The infringement theory rests on a specific sequence of software analysis steps (filtering, identifying, determining, comparing). A key factual question will be whether discovery reveals that the PalletAI software actually performs these discrete steps in the claimed order, particularly the step of first "determining a pallet design" and then comparing elements to a database based on that design. The complaint provides a screenshot from a YouTube video showing the system's GUI, which may provide some evidence of how data is processed and displayed (Compl. ¶206).
V. Key Claim Terms for Construction
Term 1: "recipe of repair operations" (’668 Patent, Claim 3)
- Context and Importance: This term is critical because it defines the required output of the claimed method. The dispute may turn on whether the PalletAI system's output—which the complaint alleges includes identifying repair requirements and sorting pallets—qualifies as a "recipe." Practitioners may focus on this term because the patent links the "recipe" directly to "machine control within the repair cell."
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim language itself does not specify the format or content of the "recipe," which may support an argument that any set of instructions leading to repair, including a simple sort command, falls within its scope.
- Evidence for a Narrower Interpretation: The specification states, "This recipe is handled by a combination of the Human Machine Interface (HMI) system, the analysis system, the programmable logic controller (PLC) and the robot controller" ('668 Patent, col. 6:5-10). This language suggests the "recipe" is a specific, integrated set of machine-readable commands for controlling an automated repair cell, not merely a high-level list of defects or a sorting decision.
Term 2: "analyzing the reflected light based on" (’360 Patent, Claim 9)
- Context and Importance: This term introduces a sequence of four mandatory analytical sub-steps. The infringement case depends on the accused system performing all four. Practitioners may focus on this term because the ordered nature of the sub-steps (e.g., "determining a pallet design" before "comparing each element") may create a high bar for infringement.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The patent states that "The analysis simplifies the data and relates the data to known facts so that a repair process... can be specified" ('360 Patent, col. 3:44-47). This could suggest that any analytical process achieving that end result meets the claim, regardless of the precise internal software architecture.
- Evidence for a Narrower Interpretation: Claim 9 explicitly lists four distinct sub-steps linked by "and," suggesting they are all required. The detailed description explains these as discrete logical processes: first, coordinates are used to identify elements; second, the "pallet design is determined by the number, size, and location of the elements;" third, "The elements are analysed against specific criteria for the pallet's determined design" ('360 Patent, col. 2:58-63). This supports an interpretation that the "design" must be determined from the scanned elements before a comparison to a database can occur.
VI. Other Allegations
- Indirect Infringement: The complaint alleges inducement of infringement based on Defendant’s marketing materials, product literature, and technical support offered to customers (Compl. ¶135, ¶226, ¶299). It is alleged that Defendant and its partner, Alliance, collaborate with customers like Niagara Pallet and iGPS on the design, installation, and maintenance of infringing systems, thereby instructing and encouraging their use (Compl. ¶143-144, ¶307-308). The complaint also alleges contributory infringement, asserting the PalletAI system is a material component of the claimed inventions with no substantial non-infringing use (Compl. ¶159, ¶250, ¶323).
- Willful Infringement: Willfulness is alleged based on both pre- and post-suit knowledge. The complaint alleges Defendant had pre-suit knowledge through its partner, Alliance, which had allegedly monitored Plaintiff's patents and admitted infringement of a related patent in February 2023 (Compl. ¶87, ¶150). Willfulness is also based on Defendant's alleged continued infringing activities—such as exhibiting the PalletAI system at trade shows and installing it for customers—after being served with the original complaint in May 2025 (Compl. ¶90-92, ¶156).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of definitional scope: can the term "recipe of repair operations," which the '668 Patent links to direct machine control of a repair cell, be construed to cover the accused system's alleged output of sorting commands and defect reports?
- A second central question will be one of operational equivalence: does the software in the accused PalletAI system perform the specific, four-part analytical sequence required by Claim 9 of the '360 Patent, or does it reach a similar result through a fundamentally different logical pathway that omits one or more of the claimed steps?
- Finally, a key issue for indirect and willful infringement will be one of imputed knowledge: to what extent can the alleged pre-suit knowledge and admissions of Defendant’s partner, Alliance, be legally imputed to the Defendant to establish the requisite intent prior to the filing of the original complaint?