【正文】
is difficult to acplish in the case of bridge construction as building a bridge can be considered a oneofakind flow as the workcrew,management,site characteristics, etc. are rarely consistent between projects. Therefore, the construction progress was followed from start to end and the data collected was used to work backwards. The studies are used to create a VSM of the current, future and ideal states that are analyzed according to Figure 3. Data for the two projects were collected through frequent construction site visits by the main author. During the site visits, the onsite management was asked to give statements on the work performed and to validate the data collected. However, the main goal of the site visits was to collect quantitative data on lead times and inventory levels. Information about lead times was achieved by collecting data on delivery of the material to the construction sites (. reinforcement deliveries) and the date for which the reinforcement was mounted and fixed into the bridges. In addition, some of the waiting time for the reinforcement in between activities was measured and estimated. It was difficult to establish the waiting time for the current and future states, since at least one ton of reinforcement is lifted upon the superstructure at the same time and after that each single rebar is manually transported and fixed on the superstructure. Through this mounting and fixing procedure it bees difficult to exactly establish the waiting time for each rebar and, hence, an average total waiting time for all the reinforcement in the superstructure bees more relevant to measure and estimate. 6 Seeing the potential of workflow improvements Waste observed in the current state was moving materials around the construction site, waiting for materials, machines or instructions, rework, and interruptions of progressive work. In fact, a study of productivity at the specific bridge indicated that only about 30 % of working time was actually adding any value. Other typical wastes that were observed are presented in Table 2. By introducing reactive solutions such as a pull system, “ milk run” deliveries and improved production planning, the current state can be improved. As a consequence of the improvement, the waste of high inventory levels and long lead times can be reduced. Also moving the reinforcement inventory around the site due to other activities is reduced dramatically. In addition, cash funds being tied up in inventory are minimized and in some cases it might even be possible to earn money on interest on advance payment from the client. In the ideal state, much of the reinforcement was pulled justintime to the construction site, . it came the same day as it was supposed to be fixed into the construction (plus a two day buffer). The ideal state improved the lead time and inventory levels from current state with approximately 90 % (Table 3). Also, the manufacturing cost at site decreased significantly (68 %) in the ideal state at the expense of an increase in the procurement of the prefabricated rebar structures of approximately 30 %. Still, the overall construction cost decreased as well as construction time onsite. Discussion amp。 increased ine and at the same time maximizing return on investment and increasing cash flow (Olhager, 1993). The traditional measures of lower operational costs, lower capital costs and increased sales would lead to better revenue for the project (Mehra et al., 2021。 hence we have a suboptimized situation. By looking at the system and only by reducing all three or having status quo on two and reducing one wil