This article is based on a presentation I recently delivered at the Africa’s Big 7 Event for food professionals, held at the Gallagher Estate in South Africa from 23 – 25 June 2018.
The focus of the presentation was on how new technologies could reduce the risk of fraudulent activities in the handling of food products throughout supply chains.
2. The global scene
Globally, the technology is still in a pioneering phase with respect to supply chain traceability and visibility. Many new technologies are being evaluated through a plethora of pilot projects in many different industries. A major global driver behind this groundswell of supply chain traceability and visibility approaches is the increased emphasis on food safety and security.
Government departments and regulators globally are demanding compliance to ever-tightening legislation and regulations of supply chains. Brand owners and retailers want to see the origin of consumer products and visibility of compliance to sustainability standards. Primary producers and agri-processing organisations want to have better visibility of committed demand from exporters and retailers. Permission and permits to export plant and animal products become stricter with respect to compliance to the rules in handling diseases, contaminants, sustainability, provenance certificates, and digital compatibility of information systems.
Another particularly challenging aspect of the current state of affairs with respect to traceability and operations visibility in supply chains globally is the lack of a universal or common framework of approaches, concepts, terminology and solution crafting. Simple business solution concepts such as traceability, supply chain visibility, supply chain architecture, value chain architecture and demand-led approach often have many different interpretations. Consequently, collaborating parties and stakeholders may confuse each other and find it difficult to reach agreements across the supply chain. Communications, written and verbal, are often at a high level and heavily laden with concepts that are derived from the current technology-hyped trends. As a result, it is not easy to fathom the exact nature of the underlying business requirements and the business benefits of the technology-based solutions on offer.
3. Let’s talk general concepts
In order to converse on the issue of improved supply chain visibility to curb fraudulent product handling throughout supply chains, let’s first clarify a few general concepts.
a. Traceability and Operations Visibility in a supply chain
Figure 1: A schematic diagramme showing an example of the various constituents of an operations visibility stack. Traceability always form the basis of a synchronous audit trail stretching across a supply chain, which is then enriched with various relevant visibility data sets according to the needs of a specific supply chain.
Traceability is the backbone of an operations visibility solution. In a supply chain context it refers specifically to the monitoring and recording of the sequential item handling events that occur in the workflow within and between organisations. This is illustrated at the bottom of the operations visibility stack in Figure 1 above. Also illustrated is the repetitive nature between different organisations of a generic traceability workflow pattern.
But traceability alone is not enough visibility! There is a need inside and outside of the supply chain to have visibility on attributes of products and processes. This is often the layer in the operations visibility stack where digital sensor data (the ambitious term would be IoT (Internet of Things)) is used to provide a robust data collection mechanism. Some of these attributes are usually indicated on the labels on consumer products. If such digital sensor data is indeed used in a supply chain, the traceability audit trail of product handling events is still required to synchronise the data sets and to provide the evidence that the product items have been subjected to the conditions measured by the sensors.
Another constituent of operations visibility, often insisted on by consumers, is the compliance to agreed standards and best practices, especially sustainability standards. Monitoring and recording of the degree of compliance to specific standards from all participants in a supply chain are increasingly done electronically as part of the functionality offered by a digital operations visibility platform. However, in many cases the recording and referencing of products to compliant processes and organisations in the supply chain is confused with traceability. For example, claiming compliance to a specific sustainability standard on the packaging of a product item without having an item level traceability capability to support the claim could create a false perception. The compliance of the process or organisation to the specific sustainability standard could be quite correct, but the product item could have travelled a different route and may not have been subjected to the specific compliant process or organisation as portrayed.
The main reason, if not the only reason, for a supply chain to produce and deliver food products according to the underlying value chain, is that there are consumers, or consumer proxies, willing to buy the final product and retailers that plan and operate according to the specific market demand. Such demand-led logic drives production throughout a supply chain. External coordination is usually required throughout a supply chain to ensure appropriate levels of communication in the cascading of retailer demand (preferably in the form of a committed demand) through all intermediate products up to primary production. This kind of visibility is what minimise over or under supply in a supply chain and contributes hugely to price stability.
This discussion serves to illustrate that there is a distinct difference between traceability, as a constituent of operations visibility, and other complementary constituents that provide additional supply chain visibility, but that still require traceability in order to synchronise data sets and to provide the required evidence that product items have been subjected to the conditions monitored, recorded and claimed.
b. Solution crafting
Solution crafting refers to the creative process of contemplating new solutions from the current portfolio of available technologies, business processes and procedures, infrastructure components, and advanced statistical and mathematical techniques (the ambitious term is artificial intelligence). All of these of course happen within the dictate of the reigning policies and business models. The software engineering and the development of appropriate enterprise level, supply chain level and value chain level architectures then follow as a way to bring all these elements together into a professional design that can be coded, tested and deployed across value chains and supply chains within.
Given the multitude of input considerations at many different levels of abstraction means that a solution of this kind goes far beyond the IT system itself. This is something that many organisations do not fully appreciate and it is evident from the many cases where a traceability software system is procured on the premise that it represents the entire solution. Inevitably these systems are blamed for failing to bring the organisation to the next level of proficiency it has aimed for.
The schematic diagramme in Figure 2 below serves to clarify the concept of a traceability and operations visibility IT-based solution. Firstly, in this case it defines four layers of enablement that exist in support of each other and that has to be aligned in order to be effective and efficient in working together in the solution stack.
Secondly it suggests a top-down ranking of these layers starting at policy/business model level and working its way down to the infrastructure level. The methodology behind solution crafting also suggests making a distinction between the business architecture and the technical architecture of the solution. In practice most often the challenge is not to design the business architecture, but to analyse a given business practice into its policy/business model components and the supporting process/procedural components. Such understanding is then structured in terms of a business architecture in order to inform the design of the technical architecture.
Thirdly it is significant to note that the software system that is procured by an organisation is only representative of the hardware/software layer of the overall solution. Without making provision to plan and implement all four layers in a coordinated way the ‘as-procured’ solution may not provide the business results expected from it.
Hence, from here on all references to a traceability and operations visibility solution refer to a solution construct that contains all four layers in a design that is appropriate to the specific business ecosystem it is designed to support.
Figure 2: A schematic diagramme illustrating four layers of enablement to be found in traceability and operations visibility solution that supports an entire value chain, and all supply chains within.
c. The nature of a supply chain traceability system
Figure 3: Schematic diagramme showing three functional layers of a generic traceability system. The data capture layer is most often hosted inside an organistion and serves to monitor and record traceability events mission critically in real time. In a supply chain scenario the top two functional layers are best provisioned as a cloud-based hosted service across an entire value chain. The structured real-time interfacing and synchronisation between the local enterprise host and the cloud-based host is critical to the integrity of the traceability audit trail that stretches across the value chain, accommodating all participating supply chains.
Given the distinction made above in section 3(a) the schematic representation shown in Figure 3 above refers to the systemic logic behind the monitoring and recording of item handling events in any specific workflow environment where physical items are being handled within and between organisations in a supply chain.
Supply chain traceability logic refers to a functional stack starting from the bottom up with data motoring and capturing with respect to physical item handling events. It should however be noted that within the context of the GS1 standards suite individual service events can also be monitored and recorded as hard as a physical item handling event. Data capturing of item handling events happens mission-critically in real-time within an organisation based on a local enterprise platform capability. This environment is focused on scanning item codes (barcodes and/or RFID tags) and applying item handling event rules and validation and verification rules and workflow rules in order to ensure the integrity of data that is taken in as being representative of the physical handling of product items on the farm, the factory floor and the shop floor.
The storing, structuring and management of traceability (item handling events) and sensor data happens in a different environment to where data is captured. Here the focus is on the supply chain business rules and data is structured according to the generic logic of Receive-Transform-Dispatch with respect to both the physical item handling and the handling of the business instructions through orders, contracts and programme definitions.
Analysis and reporting from a value chain-wide platform is still not a familiar concept and many a user misunderstood the dual need for both privacy and confidentiality, on the one hand, and supply chain visibility on the other hand. The default status of data and information on a supply chain traceability and operations visibility platform is that it belongs to the organisation that puts it there and is theirs to use as an internal management tool and only the organisation and its authorised users can see the information. However, any other supply chain participant or third party service provider or stakeholder can agree with the specific organisation to share in any or all aspects of that information set. The external coordinating body then configures a query or report to enable all parties participating in the information sharing agreement to access the agreed information. The most practical way in which this is achieved is for the information agreement to be included in the on-boarding agreement of a supply chain participant to be signed at first link-up to the platform and the supply chain–wide traceability audit trail.
It should be kept in mind that although participants need information sharing agreements to share information across organisations in a supply chain, the system itself ‘see’ all information across the entire supply chain and has the ability to continuously apply business rules, perform validation and verification and generate alerts and notifications based on supply chain rules across the entire supply chain without further agreements between participants.
d. Supply chains versus value chains
Figure 4: Schematic representation showing a four stage value chain with two supply chains operating in that value chain.
Normally the distinction between a supply chain and a value chain is only of academic interest, however in the context of supply chain traceability and operations visibility it becomes of cardinal importance. As is explained later in this document, some approaches to supply chain traceability ignore the value chain concept and only focus on supply chain architectures. Others in fact design the traceability solution within a value chain architecture and then simply link up supply chains that want to adopt a structured traceability and operations visibility logic.
Figure 4 shows a four-stage value chain with two supply chains operating within. It means that there are one primary producer, two intermediate producers, two manufacturers and one retailer. In a supply chain architecture these two supply chains will operate on two different systems, whilst in a value chain architecture they both operate on the same system. Cross-linkages in supply chains make supply chain architecture complex, especially for the retailer. Value chain architectures are better equipped to handle such cross-linkages in supply chains and provide a single interface for a retailer to all supply chains that feed into itself.
e. Validation, Verification and Authentication
Figure 5: Validation is illustrated as the act of confirming that the code used on a product item has indeed be issued by the central code management database, and that it is still active.
Validation is an automated action in a traceability and operations visibility system that checks each scan of a product item instance unique identifier code, embedded in a barcode format or a RFID tag, to ensure that it is a legitimate code issued by the central code management database of the platform and that it is still active. The traceability and operations visibility system can be configured to change the status of the unique item instance identifier code to ‘inactive’ once it passed the point-of-sale terminal at a retailer. From that point onwards the validation function will generate an alert if the same unique item instance identifier code is ‘seen’ on a product item by the system anywhere in the supply chain. In a value chain architecture this checking procedure happens against all item instance codes issued across all supply chains in that value chain.
Figure 5: Verification is illustrated as the act of confirming that the code seen on a product item has indeed been seen at the previous participant in the supply chain, and that it relates to intermediate products upstream in the supply chain up to primary production.
Verification is a procedure that is built into the logic of a supply chain traceability and operations visibility system at specific item handling events to confirm that an item has previously been handled earlier on in the supply chain, starting at the very previous (dispatching) participant from which an item was received. This procedure could potentially include any or the entire accumulated history file of a specific item instance. It means that a product without prior history on the system will trigger an alert on the system indicating it as a suspect item that requires further investigation.
Figure 6: Authentication is illustrated against a covert feature of the product item, in this case using artificial DNA (single strand) applied to the packaging, or even to the contents, of the product item. Authentication against a covert marking or an inherent unique product characteristic provides added assurance and evidence of identity.
Authentication techniques often require off-line analytical techniques to provide evidence of identity. It is normally not done in the every-day rhythm of workflow execution, but serves to provide additional evidence in cases of distrust or dispute where adjudication is required. Such markings or characteristics can be common to a specific organisation, a batch, a product line or specific to a single item instance.
4. Identify fraud by managing item handling visibly, effective and efficient
Consider the following four-stage value chain:
In order to participate in a supply chain traceability audit trail each participant need to be traceable in itself. Such traceability is derived from the following generic internal visibility logic:
Although simple in terms of the construct, it takes a highly disciplined environment to ensure that the physical procedures and the data captured about the physical procedures are perfectly aligned. This is true for all roles/stages in the value chain.
The next step is to ensure that the item handling procedures between participants in the supply chain are visible and traceable, as is shown next:
This generic logic also holds true between all participants in a supply chain, irrespective of value chain role.
Seen together these internal and external traceability procedures provide an operations visibility view that could become very tight in terms of step-by-step item handling events based on the visibility resolution designed and implemented (i.e. batch level, product line level, product instance level). Specifically it leads to elimination of avoidable errors, real time record keeping, identity validation and verification, immediate problem identification and solving, increased productivity, compliance to standard/process/procedure and provides visibility of immediate upstream and downstream activities. The increased operational visibility and traceability provide a sound basis for improved managerial decision-making and control.
However, these are the same qualities that enable an organisation to identify non-compliant processes and procedures and fraudulent actions. Fraudulent products enter the supply chain where the controls are not so tight, where information is not scrutinised at an appropriate level of granularity, and where non-real time systems are used where information can be uploaded or edited after the fact (the item handling event).
In conclusion, therefore, the visibility offered by an appropriately designed and implemented traceability and operations visibility system provides the forensic hardness and the enhanced controls required to significantly reduce, and even eliminate, the occurrence of fraud at the level of physical item handling events.
5. Different approaches to supply chain visibility
a. Honesty approaches
These approaches to supply chain traceability are based on declarations made by the participating organisations with respect to receiving raw materials, transforming raw materials into finished products, and dispatching finished products to the next supply chain participant. This approach can be illustrated as follows:
By stringing these declarations together in a hosted service environment a sense of a traceability audit trail is created that is by no means forensically hard. In order to provide a measure of trust in this approach physical on-site inspections and audits are conducted on a sampling basis.
b. Batch and quantity accounting
This approach to supply chain traceability is a well established practice between manufacturers and retailers and it relates to the principle of tracking one up and one done down relative to a participating organisation collaborating in a supply chain. It means that in itself it is not providing traceability across an entire supply chain, but it can create the links from one organisation to another, which creates the entire supply chain audit trail when added together. This approach can be illustrated as follows:
In terms of this process a manufacturer produces finished goods from raw materials. Each raw material type and finished product line is identified by means of a company specific internal stock code. Batch and/or lot numbers, via internal works orders, controls the production step. Hence for each production run the finished goods are related to the raw materials via the specific batch number. Retailers order these finished goods in terms of stock code number, batch number and quantity. In similar vein, when these goods are sold the point-of-sale system records the stock code, batch number and quantity of goods sold in order to replenish stock or simply to reconcile stock levels.
This approach is robust in terms of its traceability at batch level, although it does not convey much operations visibility information. Shown in the illustration is only the manufacturer and the retailer. However, it could be expanded to also show the interaction between the manufacturer and its suppliers.
c. Item level traceability approaches: Hosted service
In this approach a hosted environment is created to which each supply chain participant is uploading traceability data. From here various queries and reports are created for purposes of supply chain management.
Data integrity of this approach depends largely on the way in which data capture is done (e.g. automated versus manual) as well as the degree of validation and verification that is built into the system.
d. Item level traceability approaches: Distributed service
In this approach to item level traceability and operations visibility each participant in the supply chain is interfaced to the next by means of machine-to-machine data exchange interfaces. In the illustration below it is shown that company A would prepare a file with the entire history of the shipment dispatched to company B. Company B in turn would perform its transformation service to these materials and prepare a similar file on the history of the finished goods it is going to dispatch to company C. This history now includes both the history of intermediate products in company A plus that of company B.
This procedure continues until the last participant in the supply chain is reached. The obvious limitation is that only the last participant has full visibility across the entire supply chain. Also, this approach requires considerable IT skills in the participating companies in order to develop, implement and maintain these interface protocols. This procedure has to be repeated inside each supply chain
e. Distributed ledger technology (DLT) approaches
These approaches are most often referred to as blockchain applications. However, blockchain refers to the specific implementation of distributed ledger technologies in support of crypto currencies. Hence for the purpose of traceability and operations visibility this document will refer to DLT’s.
Distributed ledger technology refers to an electronic file management protocol that does the following:
- Package information about item handling events, product items and production processes into electronic files
- Linking these files into a single group of files
- Encrypt these files
- Distribute these files to predetermined users
- Can also keep the information centrally for the purpose of securely verifying a product item.
This approach is illustrated as follows:
This protocol provides a very secure and immutable environment in which the traceability information is packaged and distributed. However, as file management protocols, DLT’s do not touch the physical world of scanning a barcode label or a RFID tag. As a result the technology does not exert influence on what information nor the origin of the information that is packaged into the group of distributed ledgers. The DLT approaches therefore are well positioned to provide a verification service to confirm the identity and history of product items, provided that the information in the distributed ledger are to be trusted.
In general therefore it is apparent that a DLT-based traceability and operations visibility solution requires elements of traditional traceability systems with respect to data capture technologies to ensure that the integrity of the data in the distributed ledgers grouping is indeed trustworthy. The encryption and the strict file management protocols are also seen as potentially making it harder and more expensive to share information in a supply chain.
f. Fractal supply chain traceability
This approach is making use of a central hosted service that is setup to serve an entire value chain globally. Once setup it is deployed globally and then supports an unlimited number of supply chains anywhere in the world, provided there is reliable Internet connectivity.
This approach is illustrated as follows:
The starting point of the fractal traceability logic is an agreed (pre-determined) minimum dataset for each value chain role. This data set covers both the traceability of physical item handling events as well as the business dictates of the order/contract/programme hierarchies. These data sets represent the fractal units that are uploaded into the hosted service environment where they are then linked up according to how supply chain participants trade with each other. In this fashion the supply chains are mapped as a fractal pattern. This fractal pattern is created per product item per participant per supply chain. These fractal patterns represent the traceability audit trails that stretch across each supply chain in the value chain. At large scale this approach holds promise to become the most affordable way to create forensically hard traceability audit trails across all supply chains within a global value chain.
The data capture component of the fractal traceability approach happens within the existing systems being used at enterprise level within organisations. For those prospective participants that do not have adequate internal traceability and data capture systems to participate in the supply chain traceability audit trail, especially at farmer level, data capture screens can be exposed from the hosted service.
In order to illustrate the difference between a supply chain traceability protocol and a DLT protocol, the following comparison can be made on the basis of the required solution stack discussed above in section 2:
This illustration shows that a supply chain traceability protocol addresses the entire solution stack, whilst a DLT-based protocol (generally referred to as Blockchain, see section 5 (e)) only addresses electronic file management inside the software component of the solution stack.
This document has set out to create a common framework within which to converse about supply chain traceability and operations visibility concepts and provide the basis for the relationship between improved supply chain operations visibility and the reduction of fraud within the physical item handling event environments of supply chains.
A few important factors that have not been mentioned as yet, and which will not be discussed in this article, but that need to be considered in the implementation of traceability and operations visibility platforms across supply chains are the following:
a. Ownership of the system
Since the platform operates across one or many supply chains, it is not obvious who should own it. Experience to date shows that the investor wants to own it. However, the platform empowers and advances entire supply chains and eventually an entire industry. It does not make sense to run such a platform as a profit making enterprise. Rather it should be managed by a Non-Profit organisastion on a commercial basis under a Service Level Agreement in order to ensure maintenance, sustainability and regular updates to keep it relevant to its users.
b. Supply chain management
The enhanced visibility that is provided by supply chain traceability and operations visibility platforms is not necessarily leveraged upon by participating organisations since the existing supply chain management models do not make provision for managing suppliers and clients beyond the first, and maybe the second, tier. Much more research and experimentation is required to equip companies to leverage on supply chain visibility. Existing approaches to business models are also restricting the minds of managers in breaking out of the mould.
c. Supporting a brand promise
The traceability and operations visibility platforms are ideally suited for brand owners to obtain all the evidence they need to support their brand promise to consumers. Therefore brand owners may become a strong pressure group in future supply chains onto participant organisations to link-up to such a platform and to conform to the supply chain-wide agreed principles.
d. Sustainability and traceability
As argued above in section 3 (a) these two concepts are two sides of the same coin and sustainability of a product item cannot be claimed without a traceability platform that provide the evidence that can link the product item to the complaint process or organisation.
e. Communication to consumers
Not discussed in this article at all, but very relevant to the conversation, this matter will be discussed in a separate article. A guiding principle is that consumers should not be linked up to the same platform within which the supply chain participants are interacting. Yet, the relevant information needs to flow from the visibility platform to the platform where consumers have access. This consumer facing platform will be a powerful relationship building environment that need to be thought of carefully.
f. Closing remarks
- Fraud detection in supply chains is an outcome of well-structured operations visibility and traceability
- Traceability alone is not enough to justify your investment, extend it to full operations visibility
- If your solution does not cover all three obligations of traceability, perhaps it is not a traceability system
- Think of traceability/visibility as a business solution covering your business case/model, process and procedure, hardware and software and infrastructure
- Be bold: Validate, verify and authenticate,…otherwise your data integrity and claims will be suspect