2020Case16Mercury

Contents Lean Construction Ireland Annual Book of Cases 2020 Case 16 • Quotations from suppliers went to people outside of Pro- curement and were only provided to Procurement when the material was being requisitioned for purchase. This left zero time to adequately go to market for alternative bids. • Most quotations contained either a sea freight or an air freight adder, and which was approved by someone outside of Procurement. Airfreight alone accounted for 2.25% of total project spend. • Strong relationships were built with the suppliers and it was hard for people to change from the practices and peo- ple that they were used to. • Some suppliers provided on-site assistance to junior en- gineers with MTOs, which therefore reduced competition. • There wasn’t enough information available to enable proactive procurement. • Requisitions mostly contained free text which gave very little detail about the product being requested. POs were processed based on quotations provided by people from outside of Procurement. • In the case of one supplier, most of the products ordered could have been purchased from official distributors locally. The most significant finding out of this RCA was that, given the repetitive nature of the equipment, we had no agreed materials list and no agreed pricing structure with vendors. Data capture in our ERP system was not being maximised to enable an efficient and standardised approach to purchas- ing regularly bought items. We deployed the DMAIC method to set out our plan in terms of understanding more about the issues and solving them. The objective was to improve the buying process to a point where all repetitive materials could be bought quickly, competitively, and to the correct specification. The focus moved to having the right information, and to having it early and consistently. The need for having robust and standardised information, readily available to key users, formed the basis of thinking over the ensuing years. In fact, it became mostly about how to manage the information and capture the critical data. Defining the problem Taking into consideration the findings of our initial review of vendor spend, the extended analysis of high value suppliers found that: • Where procurement had an opportunity of increased early involvement, data was not being captured in the ERP system that would empower the team to be more proactive in the RFx process. The symptom in this instance was that POs were being placed with the same suppliers based on historical purchases. • In scenarios where RFQs were being issued to market by procurement, they were being repetitively issued for the same products. • RFQs were being issued post-requisition approval in all cases, and there was zero long-form RFQ for bulk price leveraging. • The procurement function was consumed with processing POs, with little or no time available to agree long-term conditions with vendors or to think and act strategically. • 100% of RFQs issued were in short-form outputted from the ERP. Short-form RFQ from the ERP does not ensure all terms and conditions (T&Cs) are captured in the tender process and only focusses on the price of a material. • Much wasted time was subsequently spent processing invoices. Lack of accurate information and verbal instruction were the main root causes for misaligned invoices. Measuring and analysing the issues Our first task was to examine the data available in the ERP system and understand the reasons why this data was not being used in the initial phases of procurement. The main findings were that: • Most of the data on the ERP system was either out of date or inaccurate. • Engineers and construction people spent too much time searching for information on the ERP and they could not rely on it because it was not clear. • It took too long to get new material data set up on the system. • The people responsible for raising requisitions were not adequately trained to do so. • On further analysis it was shown that the procurement team spent up to 50% of their time resolving queries that were raised by vendors due to the inadequate information provided on RFQs. Each material data record in the ERP is known as a “material master” (MM). In mid-2013, there were approximately 30,000 MM records in the ERP, and findings included: • Only 2,500 records could be used with a level of accu- racy that would enable a piece of equipment or material to be purchased using that data alone. • Only 45% of POs contained MM. • Of the 2,500 MM that could be used, for every PO that was raised, a short-form RFQ was also raised. • Zero agreed prices locked into our ERP system. • Approximately 50% of RFQs were being issued to the market repetitively for the same material. Improving the situation Our objectives started to form organically as we worked through analysing the problems. At a high level, these objectives were to: • Create time for experienced procurement people to work strategically by removing constraints such as administrative buying. • Build a data library that was robust enough to encourage engineering and construction to use it in the early phases of planning and material requisition. • Forge relationships internally that would assist in making things happen. • Forge relationships with key vendors who would see the benefits of standardising how we interacted with them. • Create SOPs and train users to be experts in the new processes.