A warehouse as Swarm of devices.

Introduction

The objective of this entry to see how we can view warehouse implementation differently and explore an implementation and its advantages

What is a warehouse?

Physical

Warehouse is a intermediate place in a supply/demand chain where goods or merchandise are stored; a storehouse. A warehouse is normally configured as a place where goods are received, stored and delivered. So it is important that this velocity of product from input door to the output door is managed effectively and efficiently. As shown in the pic, warehouse consists of equipment to store goods (a location) and equipment to move goods from one location to another. Each of these location of goods or bins are named and parts in these are recorded. When a order to deliver is received in a warehouse, it is picked from the location and is placed in a pre-designated location (could be a trailer) from where it is delivered to the end customer.

Information Perspective

So from a information perspective, the warehouse is a facility, which provides spaces into which goods can be stored. These spaces are named so it can be identified uniquely in a warehouse. The named places and the men/automatons (also named) are organized in a very coordinated manner using picking or put away strategies. The putaway strategy moves product into places in warehouse and picking strategies moves product from named places. These strategies are organized to achieve high product velocity from inbound to outbound door. This can be done by either functional or flow perspective

Modeling approaches?

Functional Modelling Currently most warehouse information systems are organized from a functional perspective. I.e. the warehouse is a collection of move functions (being executed by people) of moving product from named place to another named place. These move functions, triggered by either a delivery,receipt or transfer order, are done concurrently and are consolidated by placing them in a designated named consolidation place. These moves are coordinated using the optimum picking and putaway strategies so that the people/automatons are used efficiently or effectively. The WMS is modelled by modeling this function as computer programs, with an option to configure this into different coded put away/picking strategies

Flow Modeling We can also view the warehouse from another information point of view. I.e a point of view of flow of product. I.e instead of modeling Warehouse Management system (WMS) as product moves, model it as product flow in a warehouse. I.e say there is a product that has to be moved from inbound dock to storage location and onto the outbound delivery dock. Instead of viewing it as a function of moves recorded and sequenced in a computer, view the warehouse as a community of devices each representing itself autonomously. A product representing itself through RFID, a named place representing itself as a network aware small computer and all connected by a process/community called warehouse

So instead of modeling it based on moves it is possible to model the warehouse as interactions of these community of devices. For e.g Physical Product is represented as a capability of it to interact with all actors in a warehouse community with the behaviour of put-away followed by pick (i.e it will always want to sit somewhere). The bins i.e named places are devices with capability to do putaway (its self interest is to take a physical product and store). But on putaway, the product exhibits pick behaviour, besides the product is also hidden behind the place name. Since it is hidden behind the name(bin name), the picking cannot be done with out addressing the bin name. This way clearly, a product flow can be modelled. The putaway and picking can be controlled by controlling how these devices or actions are brought in proximity.

Representing a warehouse in this way allows for distributed computing, allowing the usage of smaller devices to be employed in a warehouse. This will significantly reduce the TCO and increase responsiveness (as the physical product represents directly). Modelling business this way leads to smart business leading to very responsive proactive business

Let us explore this with smart warehouse as an example in more detail in the next post.