By Dr. Beatriz Royo, Associate Professor.
As their thesis towards the Master of Engineering in Logistics and Supply Chain Management (ZLOG), three of our students have been analysing delivery delays based on identifying data misalignments of orders information coming from different systems. From this study, Candela Lloret, Camila Revelo and Walter Antaurco have proposed a notification alert system which could enable companies to respond proactively to supply chain disruptions, in order to maintain or restore on time delivery.
The work is based on a study of a company which manufactures Magnetic Resonance Imaging (MRI) equipment for medical diagnostics. This is a make/ assemble-to-order operation, characterised by high product value, long lead times (10 to 20 weeks from customer order receipt), and high levels of complexity. Standard parts and assemblies are acquired or manufactured against forecasts whilst bespoke components are only manufactured against a firm customer order – whilst this extends customer lead times, it removes finished goods inventory for those parts.
However, it was found that some 24% of orders were delayed because of inconsistencies in Confirmed Delivery Date (CDD) between the commercial and transportation systems, while 27% of orders were delays because of similar inconsistencies between commercial and factory/production systems – in all. Over half of order delays were attributable to a lack of data integration between the various systems. This negatively impacts brand image, customer loyalty and satisfaction or indeed achieving quarterly revenue targets.
The team looked for the causes of these misalignments, through a top-down supply chain mapping analysis, and also a bottom-up analysis of that 51% of delayed orders, looking at critical points in the process. In the former, there are ten ‘Z-status’ points, from Z0 (creating a purchase order in the commercial system) right through to Z10 – arrival at customer site. As these status points are arrived at, the actual date of execution is written over the originally predicted date. Unfortunately, because there are three systems and multiple teams involved, the CDD doesn’t get updated and remains as it was at the beginning of the production process.
However, the bottom-up analysis revealed that most misalignments occur after production is confirmed and the delivery process is started – in fact, most delays happen during distribution, between Z7 (product shipped from factory) and Z10.
It was possible to be more specific still. It transpired that 85.57% of misalignments were occurring in a single plant, whilst just 12 products accounted for 79.52% of the affected orders. Delay times also vary by country of destination, and by transport mode – road transport delays averaging 31 days, and maritime transport 28 days. The team conducted a risk evaluation for each activity in the distribution process. For example, since there is a time buffer allowed between picking the consignment up from the factory and receipt of the container at the port’s warehouse, delays here are of low severity, but only if they can be picked up and corrected during this period – otherwise the departure date is delayed: a high severity event. Any delay between loading and unloading of the container on the vessel, (a delay in handling, or departure, or at sea) is by definition of high severity as it is effectively not recoverable. From pickup at receiving port to receipt at the local distribution centre (DC), there is some buffer time, so these delays are of medium severity, and delay in shipping from the DC (which may represent a cumulation of other delays) to customer delivery is of course of high severity if it means that the scheduled CDD is not met.
How can this be resolved? A risk mitigation strategy requires improved end to end visibility of the supply chain, and in particular creating the ability to identify and address the data misalignments between the commercial, production, and transport systems as they arise.
Accurate supply chain visibility throughout the distribution process is critical: not only does it help identify inefficiencies within the process, but it would enable real-time tracking of orders, anticipation of risks such as the impacts of supply chain delays and disruptions on delivery dates, and resulting in timely and accurate product delivery which enhances customer satisfaction and loyalty (and therefore, profitability!)
The team has therefore proposed a notifications alert system which would not only identify inconsistencies but assess their impact and the risks incurred by not addressing them. The system would allow the company to detect orders that, without mitigating actions, will fail to meet the CDD promise to the customer. The system relies on probabilities of occurrence and an impact and risk matrix informed by machine learning techniques as the foundation for developing digital twins, which would use historical results to predict delays and model scenarios. Findings would be presented as a colour-coded dashboard displaying appropriate mitigation priorities and strategies.
For more information you can contact Beatriz Royo: [email protected]
